STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies

Diagnostic performance of artificial intelligence-aided caries detection on bitewing radiographs: a systematic review and meta-analysis

The accuracy of artificial intelligence-aided (AI) caries diagnosis can vary considerably depending on numerous factors. This review aimed to assess the diagnostic accuracy of AI models for caries detection and classification on bitewing radiographs. Publications after 2010 were screened in five databases. A customized risk of bias (RoB) assessment tool was developed and applied to the 14 articles that met the inclusion criteria out of 935 references. Dataset sizes ranged from 112 to 3686 radiographs. While 86 % of the studies reported a model with an accuracy of ≥80 %, most exhibited unclear or high risk of bias. Three studies compared the model's diagnostic performance to dentists, in which the models consistently showed higher average sensitivity. Five studies were included in a bivariate diagnostic random-effects meta-analysis for overall caries detection. The diagnostic odds ratio was 55.8 (95 % CI= 28.8 - 108.3), and the summary sensitivity and specificity were 0.87 (0.76 - 0.94) and 0.89 (0.75 - 0.960), respectively. Independent meta-analyses for dentin and enamel caries detection were conducted and showed sensitivities of 0.84 (0.80 - 0.87) and 0.71 (0.66 - 0.75), respectively. Despite the promising diagnostic performance of AI models, the lack of high-quality, adequately reported, and externally validated studies highlight current challenges and future research needs.

A Comparison of Five Algorithmic Methods and Machine Learning Pattern Recognition for Artifact Detection in Electronic Records of Five Different Vital Signs: A Retrospective Analysis

BACKGROUND

Research on electronic health record physiologic data is common, invariably including artifacts. Traditionally, these artifacts have been handled using simple filter techniques. The authors hypothesized that different artifact detection algorithms, including machine learning, may be necessary to provide optimal performance for various vital signs and clinical contexts.

METHODS

In a retrospective single-center study, intraoperative operating room and intensive care unit (ICU) electronic health record datasets including heart rate, oxygen saturation, blood pressure, temperature, and capnometry were included. All records were screened for artifacts by at least two human experts. Classical artifact detection methods (cutoff, multiples of SD [z-value], interquartile range, and local outlier factor) and a supervised learning model implementing long short-term memory neural networks were tested for each vital sign against the human expert reference dataset. For each artifact detection algorithm, sensitivity and specificity were calculated.

RESULTS

A total of 106 (53 operating room and 53 ICU) patients were randomly selected, resulting in 392,808 data points. Human experts annotated 5,167 (1.3%) data points as artifacts. The artifact detection algorithms demonstrated large variations in performance. The specificity was above 90% for all detection methods and all vital signs. The neural network showed significantly higher sensitivities than the classic methods for heart rate (ICU, 33.6%; 95% CI, 33.1 to 44.6), systolic invasive blood pressure (in both the operating room [62.2%; 95% CI, 57.5 to 71.9] and the ICU [60.7%; 95% CI, 57.3 to 71.8]), and temperature in the operating room (76.1%; 95% CI, 63.6 to 89.7). The CI for specificity overlapped for all methods. Generally, sensitivity was low, with only the z-value for oxygen saturation in the operating room reaching 88.9%. All other sensitivities were less than 80%.

CONCLUSIONS

No single artifact detection method consistently performed well across different vital signs and clinical settings. Neural networks may be a promising artifact detection method for specific vital signs.

EDITOR’S PERSPECTIVE

ISTH bleeding assessment tool and platelet function analyzer in children with mild inherited platelet function disorders

OBJECTIVES

To evaluate the diagnostic performance of platelet function analyzer (PFA) and The International Society on Thrombosis and Hemostasis bleeding-assessment-tool (ISTH-BAT) in detecting mild inherited platelet function disorders (IPFDs) in children with suspected bleeding disorders.

METHODS

Prospective single-center diagnostic study including consecutive patients <18 years with suspected bleeding disorder and performing a standardized workup for platelet function defects including ISTH-BAT, PFA, platelet aggregation testing, blood smear-based immunofluorescence, and next-generation sequencing-based genetic screening for IPFDs.

RESULTS

We studied 97 patients, of which 34 von Willebrand disease (VWD, 22 type-1, 11 type-2), 29 IPFDs (including delta-/alpha-storage pool disease, Glanzmann thrombasthenia, Hermansky-Pudlak syndrome) and 34 with no diagnosis. In a model combining PFA-adenosine diphosphate (ADP), PFA-epinephrine (EPI), and ISTH-BAT overall performance to diagnose IPFDs was low with area under the curves of 0.56 (95% CI 0.44, 0.69) compared with 0.84 (95% CI 0.76, 0.92) for VWD. Correlation of PFA-EPI/-ADP and ISTH-BAT was low with 0.25/0.39 Spearman's correlation coefficients. PFA were significantly prolonged in patients with VWD and Glanzmann thrombasthenia. ISTH-BAT-scores were only positive in severe bleeding disorders, but not in children with mild IPFDs or VWD.

CONCLUSION

Neither ISTH-BAT nor PFA or the combination of both help diagnosing mild IPFDs in children. PFA is suited to exclude severe IPFDs or VWD and is in this regard superior to ISTH-BAT in children.

Minor head injury in anticoagulated patients: performance of biomarkers S100B, NSE, GFAP, UCH-L1 and Alinity TBI in the detection of intracranial injury. A prospective observational study

OBJECTIVES

Data in literature indicate that in patients suffering a minor head injury (MHI), biomarkers serum levels could be effective to predict the absence of intracranial injury (ICI) on head CT scan. Use of these biomarkers in case of patients taking oral anticoagulants who experience MHI is very limited. We investigated biomarkers as predictors of ICI in anticoagulated patients managed in an ED.

METHODS

We conducted a single-cohort, prospective, observational study in an ED. Our structured clinical pathway included a first head CT scan, 24 h observation and a second CT scan. The outcome was delayed ICI (dICI), defined as ICI on the second CT scan after a first negative CT scan. We assessed the sensitivity (SE), specificity (SP), negative predictive value (NNV) and positive predictive value (PPV) of the biomarkers S100B, NSE, GFAP, UCH-L1 and Alinity TBI in order to identify dICI.

RESULTS

Our study population was of 234 patients with a negative first CT scan who underwent a second CT scan. The rate of dICI was 4.7 %. The NPV for the detection of dICI were respectively (IC 95 %): S100B 92.7 % (86.0-96.8 %,); ubiquitin C-terminal hydrolase-L1 (UCH-L1) 91.8 % (83.8-96.6 %); glial fibrillary protein (GFP) 100 % (83.2-100 %); TBI 100 % (66.4-100 %). The AUC for the detection of dICI was 0.407 for S100B, 0.563 for neuron-specific enolase (NSE), 0.510 for UCH-L1 and 0.720 for glial fibrillary acidic protein (GFAP), respectively.

CONCLUSIONS

The NPV of the analyzed biomarkers were high and they potentially could limit the number of head CT scan for detecting dICI in anticoagulated patients suffering MHI. GFAP and Alinity TBI seem to be effective to rule out a dCI, but future trials are needed.

Polymerase Chain Reaction on Respiratory Tract Specimens of Immunocompromised Patients to Diagnose Pneumocystis Pneumonia: A Systematic Review and Meta-analysis

BACKGROUND

This meta-analysis examines the comparative diagnostic performance of polymerase chain reaction (PCR) for the diagnosis of Pneumocystis pneumonia (PCP) on different respiratory tract samples, in both human immunodeficiency virus (HIV) and non-HIV populations.

METHODS

A total of 55 articles met inclusion criteria, including 11 434 PCR assays on respiratory specimens from 7835 patients at risk of PCP. QUADAS-2 tool indicated low risk of bias across all studies. Using a bivariate and random-effects meta-regression analysis, the diagnostic performance of PCR against the European Organisation for Research and Treatment of Cancer-Mycoses Study Group definition of proven PCP was examined.

RESULTS

Quantitative PCR (qPCR) on bronchoalveolar lavage fluid provided the highest pooled sensitivity of 98.7% (95% confidence interval [CI], 96.8%-99.5%), adequate specificity of 89.3% (95% CI, 84.4%-92.7%), negative likelihood ratio (LR-) of 0.014, and positive likelihood ratio (LR+) of 9.19. qPCR on induced sputum provided similarly high sensitivity of 99.0% (95% CI, 94.4%-99.3%) but a reduced specificity of 81.5% (95% CI, 72.1%-88.3%), LR- of 0.024, and LR+ of 5.30. qPCR on upper respiratory tract samples provided lower sensitivity of 89.2% (95% CI, 71.0%-96.5%), high specificity of 90.5% (95% CI, 80.9%-95.5%), LR- of 0.120, and LR+ of 9.34. There was no significant difference in sensitivity and specificity of PCR according to HIV status of patients.

CONCLUSIONS

On deeper respiratory tract specimens, PCR negativity can be used to confidently exclude PCP, but PCR positivity will likely require clinical interpretation to distinguish between colonization and active infection, partially dependent on the strength of the PCR signal (indicative of fungal burden), the specimen type, and patient population tested.

Data set terminology of deep learning in medicine: a historical review and recommendation

Medicine and deep learning-based artificial intelligence (AI) engineering represent two distinct fields each with decades of published history. The current rapid convergence of deep learning and medicine has led to significant advancements, yet it has also introduced ambiguity regarding data set terms common to both fields, potentially leading to miscommunication and methodological discrepancies. This narrative review aims to give historical context for these terms, accentuate the importance of clarity when these terms are used in medical deep learning contexts, and offer solutions to mitigate misunderstandings by readers from either field. Through an examination of historical documents, including articles, writing guidelines, and textbooks, this review traces the divergent evolution of terms for data sets and their impact. Initially, the discordant interpretations of the word 'validation' in medical and AI contexts are explored. We then show that in the medical field as well, terms traditionally used in the deep learning domain are becoming more common, with the data for creating models referred to as the 'training set', the data for tuning of parameters referred to as the 'validation (or tuning) set', and the data for the evaluation of models as the 'test set'. Additionally, the test sets used for model evaluation are classified into internal (random splitting, cross-validation, and leave-one-out) sets and external (temporal and geographic) sets. This review then identifies often misunderstood terms and proposes pragmatic solutions to mitigate terminological confusion in the field of deep learning in medicine. We support the accurate and standardized description of these data sets and the explicit definition of data set splitting terminologies in each publication. These are crucial methods for demonstrating the robustness and generalizability of deep learning applications in medicine. This review aspires to enhance the precision of communication, thereby fostering more effective and transparent research methodologies in this interdisciplinary field.

Evaluating Biases and Quality Issues in Intermodality Image Translation Studies for Neuroradiology: A Systematic Review

BACKGROUND

Intermodality image-to-image translation is an artificial intelligence technique for generating one technique from another.

PURPOSE

This review was designed to systematically identify and quantify biases and quality issues preventing validation and clinical application of artificial intelligence models for intermodality image-to-image translation of brain imaging.

DATA SOURCES

PubMed, Scopus, and IEEE Xplore were searched through August 2, 2023, for artificial intelligence-based image translation models of radiologic brain images.

STUDY SELECTION

This review collected 102 works published between April 2017 and August 2023.

DATA ANALYSIS

Eligible studies were evaluated for quality using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) and for bias using the Prediction model Risk Of Bias ASsessment Tool (PROBAST). Medically-focused article adherence was compared with that of engineering-focused articles overall with the Mann-Whitney U test and for each criterion using the Fisher exact test.

DATA SYNTHESIS

Median adherence to the relevant CLAIM criteria was 69% and 38% for PROBAST questions. CLAIM adherence was lower for engineering-focused articles compared with medically-focused articles (65% versus 73%, P < .001). Engineering-focused studies had higher adherence for model description criteria, and medically-focused studies had higher adherence for data set and evaluation descriptions.

LIMITATIONS

Our review is limited by the study design and model heterogeneity.

CONCLUSIONS

Nearly all studies revealed critical issues preventing clinical application, with engineering-focused studies showing higher adherence for the technical model description but significantly lower overall adherence than medically-focused studies. The pursuit of clinical application requires collaboration from both fields to improve reporting.

Diagnostic accuracy and clinical value of [68Ga]Ga-FAPI-46 PET/CT for staging patients with ovarian cancer: study protocol for a prospective clinical trial

BACKGROUND

[18F]Fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET) is recommended during diagnostic work-up for ovarian cancer; however, [18F]FDG PET has several inherent limitations. The novel oncologic PET-tracer fibroblast activation protein inhibitor (FAPI) has demonstrated promising results in multiple cancer types, including ovarian cancer, and could overcome the limitations of [18F]FDG PET; however, high-quality clinical studies are lacking. The primary objective of the present study is to compare the diagnostic accuracy of [68Ga]Ga-FAPI-46 PET/CT and [18F]FDG PET/CT in ovarian cancer patients and to investigate how this potential difference impacts staging and patient management.

METHODS AND DESIGN

Fifty consecutive ovarian cancer patients will be recruited from Aalborg University Hospital, Denmark. This study will be a single-center, prospective, exploratory clinical trial that adheres to the standards for reporting diagnostic accuracy studies (STARD). This study will be conducted under continuous Good Clinical Practice monitoring. The eligibility criteria for patients are as follows: (1) biopsy verified newly diagnosed ovarian cancer or a high risk of ovarian cancer and referred for primary staging with [18F]FDG PET/CT; and (2) resectable disease, i.e., candidate for primary debulking surgery or neoadjuvant chemotherapy followed by interval debulking surgery. All recruited study subjects will undergo [68Ga]Ga-FAPI-46 PET/CT at primary staging, before primary debulking surgery or neoadjuvant chemotherapy (Group A + B), in addition to conventional imaging (including [18F]FDG PET/CT). Study subjects in Group B will undergo an additional [68Ga]Ga-FAPI-46 PET/CT following neoadjuvant chemotherapy prior to interval debulking surgery. The results of the study-related [68Ga]Ga-FAPI-46 PET/CTs will be blinded, and treatment allocation will be based on common clinical practice in accordance with current guidelines. The histopathology of surgical specimens will serve as a reference standard. A recruitment period of 2 years is estimated; the trial is currently recruiting.

DISCUSSION

To our knowledge, this trial represents the largest, most extensive, and most meticulous prospective FAPI PET study conducted in patients with ovarian cancer thus far. This study aims to obtain a reliable estimation of the diagnostic accuracy of [68Ga]Ga-FAPI-46 PET/CT, shed light on the clinical importance of [68Ga]Ga-FAPI-46 PET/CT, and examine the potential applicability of [68Ga]Ga-FAPI-46 PET/CT for evaluating chemotherapy response.

TRIAL REGISTRATION

clinicaltrials.gov: NCT05903807, 2nd June 2023; and euclinicaltrials.eu EU CT Number: 2023-505938-98-00, authorized 11th September 2023.

Comment on "Does the 5-2-1 criteria identify patients with advanced Parkinson's disease? Real-world screening accuracy and burden of 5-2-1-positive patients in 7 countries"

The 5-2-1 criteria are intended to help general neurologists identify patients with advanced Parkinson's disease who may benefit from treatment optimisation, such as with a device-aided therapy. Although the 5-2-1 criteria claim to address an unmet need, we urge readers to cautiously interpret the results of this validation study.

Diagnostic accuracy of a sequence-specific Mtb-DNA hybridization assay in urine: a case-control study including subclinical TB cases

Tuberculosis (TB) caused by Mycobacterium tuberculosis (Mtb) remains one of the deadliest infectious diseases globally. Timely diagnosis is a key step in the management of TB patients and in the prevention of further transmission events. Current diagnostic tools are limited in these regards. There is an urgent need for new accurate non-sputum-based diagnostic tools for the detection of symptomatic as well as subclinical TB. In this study, we recruited 52 symptomatic TB patients (sputum Xpert MTB/RIF positive) and 58 household contacts to assess the accuracy of a sequence-specific hybridization assay that detects the presence of Mtb cell-free DNA in urine. Using sputum Xpert MTB/RIF as a reference test, the magnetic bead-capture assay could discriminate active TB from healthy household contacts with an overall sensitivity of 72.1% [confidence interval (CI) 0.59-0.86] and specificity of 95.5% (CI 0.90-1.02) with a positive predictive value of 93.9% and negative predictive value of 78.2%. The detection of Mtb-specific DNA in urine suggested four asymptomatic TB infection cases that were confirmed in all instances either by concomitant Xpert MTB/RIF sputum testing or by follow-up investigation raising the specificity of the index test to 100%. We conclude that sequence-specific hybridization assays on urine specimens hold promise as non-invasive tests for the detection of subclinical TB.

IMPORTANCE

There is an urgent need for a non-sputum-based diagnostic tool allowing sensitive and specific detection of all forms of tuberculosis (TB) infections. In that context, we performed a case-control study to assess the accuracy of a molecular detection method enabling the identification of cell-free DNA from Mycobacterium tuberculosis that is shed in the urine of tuberculosis patients. We present accuracy data that would fulfill the target product profile for a non-sputum test. In addition, recent epidemiological data suggested that up to 50% of individuals secreting live bacilli do not present with symptoms at the time of screening. We report, here, that the investigated index test could also detect instances of asymptomatic TB infections among household contacts.

RECIST 1.1 versus clinico-radiological response assessment for locally advanced cervical cancer: implications on interpreting survival outcomes of future trials

OBJECTIVE

To investigate differences in standard clinico-radiological evaluation versus Response Evaluation Criteria In Solid Tumors (RECIST) 1.1 for reporting survival outcomes in patients with locally advanced cervical cancer treated with chemoradiation and brachytherapy.

METHODS

Between November 2017 and March 2020, patients recruited in cervical cancer trials were identified. MRI at diagnosis and at least one follow-up imaging was mandatory. Disease-free survival and progression-free survival were determined using standard evaluation (clinical examination and symptom-directed imaging) and RECIST 1.1. Agreement between criteria was estimated using κ value. Sensitivity analysis was done to test the sensitivity, specificity, and accuracy of RECIST 1.1 in detecting response to treatment.

RESULTS

Sixty-nine eligible patients had at least one target lesion. Thirty-three patients (47.8%) had pathological lymph nodes. Of these 33 patients, RECIST 1.1 classified only 18% (6/33) as 'target nodal lesions' and the remaining nodes as 'non-target'. There were 6 (8.7%) and 8 (11.6%) patients with disease events using RECIST 1.1 and standard evaluation, respectively. The disease-free survival at 12, 18, and 24 months using RECIST 1.1 was 94.2%, 91.2%, 91.2%, and with standard evaluation was 94.2%, 89.7%, and 88.2%, respectively (p=0.58). Whereas, progression-free survival at 12, 18, and 24 months using RECIST 1.1 and standard evaluation were same (94.2%, 91.2%, and 91.2%, respectively). The κ value was 0.84, showing strong agreement in assessing disease-free survival, although an absolute difference of 3% between endpoint assessment methodologies. RECIST 1.1 had a sensitivity of 75% (95% CI 34.91% to 96.81%), specificity of 100% (95% CI 94.13% to 100%), and accuracy of 97.1% (95% CI 89.92% to 99.65%).

CONCLUSIONS

The study showed 1.5% and 3% difference in disease-free survival at 18 and 24 months and no difference in progression-free survival between RECIST 1.1 and standard evaluation in a patient cohort with low event rate.

Malnutrition screening tool and nutritional screening tool for classification of nutritional risk in patients with cancer upon hospital admission: Comparison of diagnostic performance using Global Leadership Initiative on malnutrition criteria as reference

BACKGROUND & AIMS

Tools for screening of nutrition risk in patients with cancer are usually validated against other screening instruments. Here with the performance of Malnutrition Screening Tool (MST) and Nutritional Screening Tool (NUTRISCORE) to identify the risk of malnutrition was assessed. A full nutritional evaluation and diagnosis following criteria from the Global Leadership Initiative of Malnutrition (GLIM) was the reference standard for the classification of malnutrition.

METHODS

Diagnostic test prospective analysis of adult patients with a confirmed diagnosis of cancer. MST, NUTRISCORE and nutritional evaluation and diagnosis by GLIM criteria were independently performed within 24 h of admission to a 4th tier hospital in Bogotá, Colombia.

RESULTS

From 439 patients the sensitivity and specificity of MST was 75% and 94% and of NUTRISCORE 45% and 97% respectively. The area under receiver operating characteristic (ROC) curves were 0.90 for MST and 0.85 for NUTRISCORE (p = 0.003).

CONCLUSION

The MST showed a significantly better diagnostic performance over NUTRISCORE for detection of malnutrition risk at admission to hospital of patients with cancer.

Diagnostic validity and reliability of BT-RADS in the management of recurrent high-grade glioma

BACKGROUND AND PURPOSE

BT-RADS is a new framework system for reporting the treatment response of brain tumors. The aim of the study was to assess the diagnostic performance and reliability of the BT-RADS in predicting the recurrence of high-grade glioma (HGG).

MATERIALS AND METHODS

This prospective single-center study recruited 81 cases with previously operated and pathologically proven HGG. The patients underwent baseline and follow-up contrast-enhanced MRI (CE-MRI). Two neuro-radiologists with ten years-experience in neuroimaging independently analyzed and interpreted the MRI images and assigned a BT-RADS category for each case. To assess the diagnostic accuracy of the BT-RADS for detecting recurrent HGG, the reference standard was the histopathology for BT-RADS categories 3 and 4, while neurological clinical examination and clinical follow up were used as a reference for BT-RADS categories 1 and 2. The inter-reader agreement was assessed using the Cohen's Kappa test.

RESULTS

The study included 81 cases of HGG, of which 42 were recurrent and 39 were non-recurrent HGG cases based on the reference test. BT-RADS 3B was the best cutoff for predicting recurrent HGG with a sensitivity of 90.5 % to 92.9 %, specificity of 76.9 % to 84.6 %, and accuracy of 83.9 % to 88.9 %, based on both readers. The BT-RADS showed a substantial inter-reader agreement with a K of 0.710 (P = 0.001).

CONCLUSIONS

The BT-RADS is a valid and reliable framework for predicting recurrent HGG. Moreover, BT-RADS can help neuro-oncologists make clinical decisions that can potentially improve the patient's outcome.

Ultrasound Secondary Signs for the Diagnosis of Pediatric Distal Forearm Fractures: A Diagnostic Study

OBJECTIVE

To evaluate the diagnostic accuracy of ultrasound secondary signs of fractures in pediatric patients aged 5-15 y presenting to the emergency department with a clinically non-deformed distal forearm injury.

METHODS

This diagnostic study was conducted in South East Queensland, Australia. Emergency clinicians performed point-of-care ultrasound on eligible patients and recorded secondary signs of fractures (pronator quadratus hematoma [PQH] sign, periosteal hematoma, visible angulation) or physeal fractures (fracture-to-physis distance [FPD], physis alteration). The reference standard was the final fracture diagnosis determined by expert panel. The primary outcome was the diagnostic accuracy of secondary signs for cortical breach and physeal fractures. Diagnostic statistics were reported for each relevant secondary sign.

RESULTS

A total of 135 participants were enrolled. The expert panel diagnosed 48 "no" fracture, 52 "buckle" fracture and 35 "other" fracture. All "other" fractures were cortical breach fractures and included 15 Salter-Harris II fractures. The PQH sign demonstrated high sensitivity and moderate specificity to diagnose cortical breach fractures (91%, 95% Confidence Interval [CI] 78%-97% and 82%, 73%-88%). Poor sensitivity but high specificity was observed for the visible angulation and periosteal hematoma secondary signs. FPD <1cm showed perfect sensitivity and moderate specificity (100%, 80%-100% and 85%, 78%-90%) for diagnosis of Salter-Harris II fracture. Conversely, physis alteration showed poor sensitivity but excellent specificity (40%, 20%-64% and 99%, 95%-100%) for the diagnosis of Salter-Harris II fractures.

CONCLUSION

Ultrasound secondary signs showed good diagnostic accuracy for both cortical breach fractures and Salter-Harris II fractures. Future research should consider optimal use of secondary signs to improve diagnostic accuracy.

Subtype Identification of Surgically Curable Primary Aldosteronism During Treatment With Mineralocorticoid Receptor Blockade

BACKGROUND

Current guidelines and consensus documents recommend withdrawal of mineralocorticoid receptor antagonists (MRAs) before primary aldosteronism (PA) subtyping by adrenal vein sampling (AVS), but this practice can cause severe hypokalemia and uncontrolled high blood pressure. Our aim was to investigate if unilateral PA can be identified by AVS during MRA treatment.

METHODS

We compared the rate of unilateral PA identification between patients with and without MRA treatment in large data sets of patients submitted to AVS while off renin-angiotensin system blockers and β-blockers. In sensitivity analyses, the between-group differences of lateralization index values after propensity score matching and the rate of unilateral PA identification in subgroups with undetectable (≤2 mUI/L), suppressed (<8.2 mUI/L), and unsuppressed (≥8.2 mUI/L) direct renin concentration levels were also evaluated.

RESULTS

Plasma aldosterone concentration, direct renin concentration, and blood pressure values were similar in non-MRA-treated (n=779) and MRA-treated (n=61) patients with PA, but the latter required more antihypertensive agents (P=0.001) and showed a higher rate of adrenal nodules (82% versus 67%; P=0.022) and adrenalectomy (72% versus 54%; P=0.01). However, they exhibited no significant differences in commonly used AVS indices and the area under the receiving operating characteristic curve of lateralization index, both under unstimulated conditions and postcosyntropin. Several sensitivity analyses confirmed these results in propensity score matching adjusted models and in patients with undetectable, or suppressed or unsuppressed renin levels.

CONCLUSIONS

At doses that controlled blood pressure and potassium levels, MRAs did not preclude the identification of unilateral PA at AVS.

REGISTRATION

URL: https://www.clinicaltrials.gov; Unique identifier: NCT01234220.

Protocol for conducting a systematic review on diagnostic accuracy in clinical research

In the landscape of modern medicine, the ability to accurately diagnose various clinical conditions is paramount. As new diagnostic tools continue to emerge, their accuracy must be rigorously assessed before clinical implementation. This paper introduces a systematic review protocol tailored for diagnostic accuracy studies, drawing inspiration from a review on dysphagia screening in post-stroke patients. The protocol, designed with precision and transparency at its core, facilitates a thorough synthesis of evidence, employing tools such as the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) and the Standards for Reporting of Diagnostic Accuracy Studies (STARD) checklist for robust evaluation. The protocol emphasizes registration with the PROSPERO database and adherence to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The systematic search approach encompasses a comprehensive exploration of databases and precise keyword combinations. Distinctive inclusion and exclusion criteria, coupled with a dual-reviewer methodology, ensure the selection of high-quality studies. This framework has the potential to serve as a benchmark for systematic reviews in diagnostic accuracy, highlighting the importance of standardization, transparency, and adaptability in clinical research. This approach paves the way for a research methodology that delves deeper into diagnostic tools across various clinical scenarios, promoting evidence-based advancements in patient care.

Cheeks appearance as a novel predictor of obstructive sleep apnea: the CASA score study

STUDY OBJECTIVES

Four well-established predictors of obstructive sleep apnea (OSA) risk are body mass index, age, sex, and neck circumference. We have previously reported cheeks appearance as an OSA predictor, which may represent a combination of such predictors in a single, readily available feature. This study sought to answer the question: Is cheeks appearance an OSA risk predictor?

METHODS

This was a prospective cross-sectional diagnostic accuracy study based on STARD (standards for reporting diagnostic accuracy studies). Patients undergoing polysomnography to investigate sleep complaints at a sleep clinic affiliated with a university hospital were assessed using cheeks appearance scored 0-3 for volume and 0-3 for flaccidity to create the Cheeks Appearance for Sleep Apnea (CASA) score ranging from 0 to 6. Appearance was judged by 3 blinded and independent evaluators.

RESULTS

Among 265 patients evaluated, 248 were included. Fifty-seven patients had a CASA score of 0 and 191 had a CASA score between 1 and 6. Polysomnography diagnosed 177 of the individuals with OSA; of these, 167 had an altered CASA score. Sensitivity was 87%, specificity was 82%, positive-predictive value was 94%, negative-predictive value was 66%, and accuracy was 86%.

CONCLUSIONS

Our results suggest that combining volume and flaccidity of cheeks appearance in a single index may constitute a reliable OSA predictor. CASA score is a novel predictor of OSA with internal validity in a sleep laboratory adult population. Our findings support further studies to confirm the external validity of this practical diagnostic tool.

CLINICAL TRIAL REGISTRATION

Registry: ClinicalTrials.gov; Name: Cheeks Appearance as a Novel Predictor of Obstructive Sleep Apnea: The CASA Score Study (CASA); URL: https://clinicaltrials.gov/study/NCT04980586; Identifier: NCT04980586.

CITATION

Prikladnicki A, Gomes E, Sousa LCCR, Gonçalves SC, Martinez D. Cheeks appearance as a novel predictor of obstructive sleep apnea: the CASA score study. J Clin Sleep Med. 2024;20(6):879-885.

Transfer Learning-Based B-Line Assessment of Lung Ultrasound for Acute Heart Failure

OBJECTIVE

B-lines assessed by lung ultrasound (LUS) outperform physical exam, chest radiograph, and biomarkers for the associated diagnosis of acute heart failure (AHF) in the emergent setting. The use of LUS is however limited to trained professionals and suffers from interpretation variability. The objective was to utilize transfer learning to create an AI-enabled software that can aid novice users to automate LUS B-line interpretation.

METHODS

Data from an observational AHF LUS study provided standardized cine clips for AI model development and evaluation. A total of 49,952 LUS frames from 30 patients were hand scored and trained on a convolutional neural network (CNN) to interpret B-lines at the frame level. A random independent evaluation set of 476 LUS clips from 60 unique patients assessed model performance. The AI models scored the clips on both a binary and ordinal 0-4 multiclass assessment.

RESULTS

A multiclassification AI algorithm had the best performance at the binary level when applied to the independent evaluation set, AUC of 0.967 (95% CI 0.965-0.970) for detecting pathologic conditions. When compared to expert blinded reviewer, the 0-4 multiclassification AI algorithm scale had a reported linear weighted kappa of 0.839 (95% CI 0.804-0.871).

CONCLUSIONS

The multiclassification AI algorithm is a robust and well performing model at both binary and ordinal multiclass B-line evaluation. This algorithm has the potential to be integrated into clinical workflows to assist users with quantitative and objective B-line assessment for evaluation of AHF.

Changes in vessel density patterns assessed with OCTA in patients with diabetic macular edema treated with anti-VEGF therapy

AIMS

To determine the presence of sectoral changes in vessel density (VD) patterns induced by vascular endothelial growth factor inhibitors (anti-VEGF) in patients with diabetic macular edema (DME) using optical coherence tomography angiography (OCTA).

METHODS

Prospective, interventional study. A total of 43 patients (63 eyes) were initially enrolled in the study. We performed swept source (SS) OCT and sectorial OCTA measurement to determine parafoveal VD at baseline and after six months of anti-VEGF treatment. In the locations with statistically significant differences in VD between baseline and month 6, we performed univariate and multivariate analyses to determine which, if any, of the baseline variables were associated with the observed changes.

RESULTS

A total of 34 patients (48 eyes) were included in the final analysis. Mean VD decreased from baseline to month 6 (from 45.2 (± 3.5) to 44.6 (± 3.2) % in the SCP and from 50 (± 3.3) to 49 (± 3.9) % in the DCP). The only significant changes in VD were observed in the nasal sector of the deep capillary plexus, with a decrease of 2.9% (p = 0.001). On univariate and multivariate analyses, the only variable significantly associated with changes in VD in the nasal sector after 6 months of treatment was baseline VD in the same sector.

CONCLUSIONS

Anti-VEGF therapy has a small impact on VD values over time. These variations observed after treatment seems to be related to changes over areas of vascular anomalies and displaced vessels adjacent to cystic areas, with no significant changes over ischemic areas. No correlation was observed between this trend and other clinical baseline features.

Accuracy of optical diagnosis with narrow band imaging in the surveillance of ulcerative colitis: a prospective study comparing Kudo, Kudo-IBD and NICE classifications

PURPOSE

The diagnostic accuracy of Narrow Band Imaging (NBI) in the endoscopic surveillance of ulcerative colitis (UC) has been disappointing in most trials which used the Kudo classification. We aim to compare the performance of NBI in the lesion characterization of UC, when applied according to three different classifications (NICE, Kudo, Kudo-IBD).

METHODS

In a prospective, real-life study, all visible lesions found during consecutive surveillance colonoscopies with NBI (Exera-II CV-180) for UC were classified as suspected or non-suspected for neoplasia according to the NICE, Kudo and Kudo-IBD criteria. The sensitivity (SE), specificity (SP), positive (+LR) and negative (-LR) likelihood ratios of the three classifications were calculated, using histology as the reference standard.

RESULTS

394 lesions (mean size 6 mm, range 2-40 mm) from 84 patients were analysed. Twenty-one neoplastic (5%), 49 hyperplastic (12%), and 324 inflammatory (82%) lesions were found. The diagnostic accuracy of the NICE, Kudo and Kudo-IBD classifications were, respectively: SE 76%-71%-86%; SP 55-69%-79% (p < 0.05 Kudo-IBD vs. both Kudo and NICE); +LR 1.69-2.34-4.15 (p < 0.05 Kudo-IBD vs. both Kudo and NICE); -LR 0.43-0.41-0.18.

CONCLUSION

The diagnostic accuracy of NBI in the differentiation of neoplastic and non-neoplastic lesions in UC is low if used with conventional classifications of the general population, but it is significantly better with the modified Kudo classification specific for UC.

European Society of Cardiology 0/1-hour algorithm (high-sensitivity cardiac troponin T) performance across distinct age groups

BACKGROUND

To determine if the European Society of Cardiology 0/1-hour (ESC 0/1-h) algorithm with high-sensitivity cardiac troponin T (hs-cTnT) meets the ≥99% negative predictive value (NPV) safety threshold for 30-day cardiac death or myocardial infarction (MI) in older, middle-aged and young subgroups.

METHODS

We conducted a subgroup analysis of adult emergency department patients with chest pain prospectively enrolled from eight US sites (January 2017 to September 2018). Patients were stratified into rule-out, observation and rule-in zones using the hs-cTnT ESC 0/1-h algorithm and classified as older (≥65 years), middle aged (46-64 years) or young (21-45 years). Patients had 0-hour and 1-hour hs-cTnT measures (Roche Diagnostics) and a History, ECG, Age, Risk factor and Troponin (HEART) score. Fisher's exact tests compared rule-out and 30-day cardiac death or MI rates between ages. NPVs with 95% CIs were calculated for the ESC 0/1-h algorithm with and without the HEART score.

RESULTS

Of 1430 participants, 26.9% (385/1430) were older, 57.4% (821/1430) middle aged and 15.7% (224/1430) young. Cardiac death or MI at 30 days occurred in 12.8% (183/1430). ESC 0/1-h algorithm ruled out 35.6% (137/385) of older, 62.1% (510/821) of middle-aged and 79.9% of (179/224) young patients (p<0.001). NPV for 30-day cardiac death or MI was 97.1% (95% CI 92.7% to 99.2%) among older patients, 98.4% (95% CI 96.9% to 99.3%) in middle-aged patients and 99.4% (95% CI 96.9% to 100%) among young patients. Adding a HEART score increased NPV to 100% (95% CI 87.7% to 100%) for older, 99.2% (95% CI 97.2% to 99.9%) for middle-aged and 99.4% (95% CI 96.6% to 100%) for young patients.

CONCLUSIONS

In older and middle-aged adults, the hs-cTnT ESC 0/1-h algorithm was unable to reach a 99% NPV for 30-day cardiac death or MI unless combined with a HEART score.

TRIAL REGISTRATION NUMBER

NCT02984436.

Artificial intelligence enables precision diagnosis of cervical cytology grades and cervical cancer

Cervical cancer is a significant global health issue, its prevalence and prognosis highlighting the importance of early screening for effective prevention. This research aimed to create and validate an artificial intelligence cervical cancer screening (AICCS) system for grading cervical cytology. The AICCS system was trained and validated using various datasets, including retrospective, prospective, and randomized observational trial data, involving a total of 16,056 participants. It utilized two artificial intelligence (AI) models: one for detecting cells at the patch-level and another for classifying whole-slide image (WSIs). The AICCS consistently showed high accuracy in predicting cytology grades across different datasets. In the prospective assessment, it achieved an area under curve (AUC) of 0.947, a sensitivity of 0.946, a specificity of 0.890, and an accuracy of 0.892. Remarkably, the randomized observational trial revealed that the AICCS-assisted cytopathologists had a significantly higher AUC, specificity, and accuracy than cytopathologists alone, with a notable 13.3% enhancement in sensitivity. Thus, AICCS holds promise as an additional tool for accurate and efficient cervical cancer screening.

The accuracy of prehospital triage decisions in English trauma networks - a case-cohort study

BACKGROUND

Care for injured patients in England is provided by inclusive regional trauma networks. Ambulance services use triage tools to identify patients with major trauma who would benefit from expedited Major Trauma Centre (MTC) care. However, there has been no investigation of triage performance, despite its role in ensuring effective and efficient MTC care. This study aimed to investigate the accuracy of prehospital major trauma triage in representative English trauma networks.

METHODS

A diagnostic case-cohort study was performed between November 2019 and February 2020 in 4 English regional trauma networks as part of the Major Trauma Triage Study (MATTS). Consecutive patients with acute injury presenting to participating ambulance services were included, together with all reference standard positive cases, and matched to data from the English national major trauma database. The index test was prehospital provider triage decision making, with a positive result defined as patient transport with a pre-alert call to the MTC. The primary reference standard was a consensus definition of serious injury that would benefit from expedited major trauma centre care. Secondary analyses explored different reference standards and compared theoretical triage tool accuracy to real-life triage decisions.

RESULTS

The complete-case case-cohort sample consisted of 2,757 patients, including 959 primary reference standard positive patients. The prevalence of major trauma meeting the primary reference standard definition was 3.1% (n=54/1,722, 95% CI 2.3 - 4.0). Observed prehospital provider triage decisions demonstrated overall sensitivity of 46.7% (n=446/959, 95% CI 43.5-49.9) and specificity of 94.5% (n=1,703/1,798, 95% CI 93.4-95.6) for the primary reference standard. There was a clear trend of decreasing sensitivity and increasing specificity from younger to older age groups. Prehospital provider triage decisions commonly differed from the theoretical triage tool result, with ambulance service clinician judgement resulting in higher specificity.

CONCLUSIONS

Prehospital decision making for injured patients in English trauma networks demonstrated high specificity and low sensitivity, consistent with the targets for cost-effective triage defined in previous economic evaluations. Actual triage decisions differed from theoretical triage tool results, with a decreasing sensitivity and increasing specificity from younger to older ages.

Acute and Long-Term Management of Blunt Cerebrovascular Injury at a Quaternary Referral Level 1 Trauma Center: The Memphis Experience and Comparison of a New Scoring System

BACKGROUND AND OBJECTIVES

The management of blunt cerebrovascular injuries (BCVIs) remains an important topic within trauma and neurosurgery today. There remains a lack of consensus within the literature and significant variation across institutions. The purpose of this study was to evaluate management of BCVI at a large, tertiary referral trauma center.

METHODS

Institutional Review Board approval was obtained to conduct a retrospective review of patients with BCVI at our Level 1 Trauma Center. Computed tomography angiography was used to identify BCVI for each patient. Patient information was collected, and statistical analysis was performed. With the included risk factors for ischemic complications, a novel scoring system based on ischemic risk, the "Memphis Score," was developed and evaluated to grade BCVI.

RESULTS

Two hundred seventeen patients with BCVI from July 2020 to August 2022 were identified. The most common mechanism of injury was motor vehicle collision (141, 65.0%). Vertebral arteries were the most common vessel injured (136, 51.1%) with most injuries occurring at a high cervical location (101, 38.0%). Denver Grade 1 injuries (89, 33.5%) and a Memphis Score of 1 were most frequent (172, 64.6%), and initial anticoagulation with heparin drip was initiated 56.7% of the time (123). Endovascular treatment was required in 24 patients (11.1%) and was usually performed in the first 48 hours (15, 62.5%). While Denver Grade (P = .019) and Memphis Score (P < .00001) were significantly higher in those patients undergoing endovascular treatment, only the Memphis Score demonstrated a significant difference between those patients who had stroke or worsening on follow-up imaging and those who did not (P = .0009).

CONCLUSION

Although BCVI management has improved since early investigative efforts, institutions must evaluate and share their data to help clarify outcomes. The novel "Memphis Score" presents a standardized framework to communicate ischemic risk and guide management of BCVI.

Moving Forward on Tumor Pathology Research Reporting: A Guide for Pathologists From the World Health Organization Classification of Tumors Living Evidence Gap Map by Tumour Type Group

Evidence-based medicine (EBM) can be an unfamiliar territory for those working in tumor pathology research, and there is a great deal of uncertainty about how to undertake an EBM approach to planning and reporting histopathology-based studies. In this article, reviewed and endorsed by the Word Health Organization International Agency for Research on Cancer's International Collaboration for Cancer Classification and Research, we aim to help pathologists and researchers understand the basics of planning an evidence-based tumor pathology research study, as well as our recommendations on how to report the findings from these. We introduce some basic EBM concepts, a framework for research questions, and thoughts on study design and emphasize the concept of reporting standards. There are many study-specific reporting guidelines available, and we provide an overview of these. However, existing reporting guidelines perhaps do not always fit tumor pathology research papers, and hence, here, we collate the key reporting data set together into one generic checklist that we think will simplify the task for pathologists. The article aims to complement our recent hierarchy of evidence for tumor pathology and glossary of evidence (study) types in tumor pathology. Together, these articles should help any researcher get to grips with the basics of EBM for planning and publishing research in tumor pathology, as well as encourage an improved standard of the reports available to us all in the literature.

Novel biomarker profiles to improve individual diagnosis and prognosis in patients with suspected inflammatory bowel disease: protocol for the Nordic inception cohort study (NORDTREAT)

INTRODUCTION

Inflammatory bowel disease (IBD), including ulcerative colitis and Crohn's disease, can be challenging to diagnose, and treatment outcomes are difficult to predict. In the NORDTREAT cohort study, a Nordic prospective multicentre study, we aim to identify novel molecular biomarkers of diagnostic value by assessing the diagnostic test accuracy (cross-sectionally), as well as the prognostic utility when used as prognostic markers in the long-term (cohort study). In the diagnostic test accuracy study, the primary outcome is a successful diagnosis using one or more novel index tests at baseline compared with the ECCO criteria as the reference standard. The composite outcome of the prognostic utility study is 'severe IBD' within 52 weeks from inclusion, defined as one or more of the following three events: IBD-related surgery, IBD-related hospitalisation or IBD-related death.

METHODS AND ANALYSIS

We aim to recruit 800 patients referred on suspicion of IBD to this longitudinal observational study, a collaboration between 11 inclusion sites in Denmark, Iceland, Norway and Sweden. Inclusion will occur from February 2022 until December 2023 with screening and baseline visits for all participants and three outcome visits at weeks 12, 26 and 52 after baseline for IBD-diagnosed patients. Biological material (blood, faeces, biopsies, urine and hair), clinical data and lifestyle information will be collected during these scheduled visits.

ETHICS AND DISSEMINATION

This study will explore novel biomarkers to improve diagnostic accuracy and prediction of disease progression, thereby improving medical therapy and the quality of life for patients with IBD.The study is approved by the Ethics Committee (DK: S-20200051, v1.4, 16.10.2021; IS: VSNb2021070006/03.01, NO: 193064; SE: DNR 2021-05090) and the Danish Data Protecting Agency (20/54594). Results will be disseminated through peer-reviewed journals, patient associations and presentations at international conferences.

CLINICAL TRIAL REGISTRATION NUMBER

NCT05414578; Pre-results.

Dynamic changes of hepatic vein Doppler velocities predict preload responsiveness in mechanically ventilated critically ill patients

BACKGROUND

Assessment of dynamic parameters to guide fluid administration is one of the mainstays of current resuscitation strategies. Each test has its own limitations, but passive leg raising (PLR) has emerged as one of the most versatile preload responsiveness tests. However, it requires real-time cardiac output (CO) measurement either through advanced monitoring devices, which are not routinely available, or echocardiography, which is not always feasible. Analysis of the hepatic vein Doppler waveform change, a simpler ultrasound-based assessment, during a dynamic test such as PLR could be useful in predicting preload responsiveness. The objective of this study was to assess the diagnostic accuracy of hepatic vein Doppler S and D-wave velocities during PLR as a predictor of preload responsiveness.

METHODS

Prospective observational study conducted in two medical-surgical ICUs in Chile. Patients in circulatory failure and connected to controlled mechanical ventilation were included from August to December 2023. A baseline ultrasound assessment of cardiac function was performed. Then, simultaneously, ultrasound measurements of hepatic vein Doppler S and D waves and cardiac output by continuous pulse contour analysis device were performed during a PLR maneuver.

RESULTS

Thirty-seven patients were analyzed. 63% of the patients were preload responsive defined by a 10% increase in CO after passive leg raising. A 20% increase in the maximum S wave velocity after PLR showed the best diagnostic accuracy with a sensitivity of 69.6% (49.1-84.4) and specificity of 92.8 (68.5-99.6) to detect preload responsiveness, with an area under curve of receiving operator characteristic (AUC-ROC) of 0.82 ± 0.07 (p = 0.001 vs. AUC-ROC of 0.5). D-wave velocities showed worse diagnostic accuracy.

CONCLUSIONS

Hepatic vein Doppler assessment emerges as a novel complementary technique with adequate predictive capacity to identify preload responsiveness in patients in mechanical ventilation and circulatory failure. This technique could become valuable in scenarios of basic hemodynamic monitoring and when echocardiography is not feasible. Future studies should confirm these results.

Diffusion-weighted magnetic resonance imaging for the diagnosis of giant cell arteritis: a comparison with T1-weighted black-blood imaging

OBJECTIVES

To investigate the diagnostic performance of diffusion-weighted imaging (DWI) of the superficial cranial arteries in the diagnosis of GCA.

METHODS

Retrospectively, 156 patients with clinically suspected GCA were included. A new 4-point ordinal DWI rating scale was developed. A post-contrast, fat-suppressed, T1-weighted 'black-blood' sequence (T1-BB) was rated for comparison. Ten arterial segments were assessed: common superficial temporal arteries, temporal and parietal branches, occipital and posterior auricular arteries bilaterally. The expert clinical diagnosis after ≥6 months of follow-up was the diagnostic reference standard. Diagnostic accuracy was evaluated for different rating methods.

RESULTS

The study cohort consisted of 87 patients with and 69 without GCA. For DWI, the area under the curve was 0.90. For a cut-off of ≥2 consecutive pathological slices, DWI showed a sensitivity of 75.9%, a specificity of 94.2% and a positive likelihood ratio of 13.09. With a cut-off of ≥3 consecutive pathological slices, sensitivity was 70.1%, specificity was 98.6% and the positive likelihood ratio was 48.38. For the T1-BB, values were 88.5%, 88.4% and 7.63, respectively. The inter-rater analysis for DWI with a cut-off of ≥2 pathological slices showed a kappa of 1.00 on the patient level and 0.85 on the arterial segment level. For the T1-BB the kappa was 0.78 and 0.79, respectively.

CONCLUSION

DWI of the superficial cranial arteries demonstrates a good diagnostic accuracy and reliability for the diagnosis of GCA. DWI is widely available and can be used immediately in clinical practice for patients with suspected GCA.

Multimodal analysis of cfDNA methylomes for early detecting esophageal squamous cell carcinoma and precancerous lesions

Detecting early-stage esophageal squamous cell carcinoma (ESCC) and precancerous lesions is critical for improving survival. Here, we conduct whole-genome bisulfite sequencing (WGBS) on 460 cfDNA samples from patients with non-metastatic ESCC or precancerous lesions and matched healthy controls. We develop an expanded multimodal analysis (EMMA) framework to simultaneously identify cfDNA methylation, copy number variants (CNVs), and fragmentation markers in cfDNA WGBS data. cfDNA methylation markers are the earliest and most sensitive, detectable in 70% of ESCCs and 50% of precancerous lesions, and associated with molecular subtypes and tumor microenvironments. CNVs and fragmentation features show high specificity but are linked to late-stage disease. EMMA significantly improves detection rates, increasing AUCs from 0.90 to 0.99, and detects 87% of ESCCs and 62% of precancerous lesions with >95% specificity in validation cohorts. Our findings demonstrate the potential of multimodal analysis of cfDNA methylome for early detection and monitoring of molecular characteristics in ESCC.

Feasibility study comparing synthesized mammography with digital breast tomosynthesis and digital mammography for simulated first round screening in a single BreastScreen NSW centre

INTRODUCTION

While digital breast tomosynthesis (DBT) has proven to enhance cancer detection and reduce recall rates (RR), its integration into BreastScreen Australia for screening has been limited, in part due to perceived cost implications. This study aims to assess the cost effectiveness of digital mammography (DM) compared with synthesized mammography and DBT (SM + DBT) in a first round screening context for short-term outcomes.

METHODS

Clients recalled for nonspecific density (NSD) as a single lesion by both readers at the Northern Sydney Central Coast BreastScreen service in 2019 were included. Prior images were excluded to simulate first-round screening. Eleven radiologists read DM and synthesized mammography with DBT (SM + DBT) images 4 weeks apart. Recall rates (RR), reading time, and diagnostic parameters were measured, and costs for screen reading and assessment were calculated.

RESULT

Among 65 clients studied, 13 were diagnosed with cancer, with concordant cancer recalls. SM + DBT reduced recall rates (RR), increased reading time, maintained cancer detection sensitivity, and significantly improved other diagnostic parameters, particularly false positive rates. Benign biopsy recalls remained equivalent. While SM + DBT screen reading cost was significantly higher than DM (DM AU$890 ± 186 vs SM + DBT AU$1279 ± 265; P < 0.001), the assessment cost (DM AU$29,504 ± 9427 vs SM + DBT AU$18,021 ± 5606; P < 0.001), and combined screen reading and assessment costs were significantly lower (DM AU$30,394 ± 9508 vs SM + DBT AU$19,300 ± 5721; P = 0.001). SM + DBT screen reading and assessment of 65 patients resulted in noteworthy cost savings (AU$11,094), equivalent to assessing 12 additional clients.

CONCLUSION

In first round screening, DBT yields significant cost savings by effectively reducing unnecessary recalls to assessment while maintaining diagnostic efficacy.

Diagnostic or prognostic? Decoding the role of embryo selection on in vitro fertilization treatment outcomes

In this review, we take a fresh look at embryo assessment and selection methods from the perspective of diagnosis and prognosis. On the basis of a systematic search in the literature, we examined the evidence on the prognostic value of different embryo assessment methods, including morphological assessment, blastocyst culture, time-lapse imaging, artificial intelligence, and preimplantation genetic testing for aneuploidy.

Artificial intelligence-based classification of breast lesion from contrast enhanced mammography: a multicenter study

PURPOSE

The authors aimed to establish an artificial intelligence (AI)-based method for preoperative diagnosis of breast lesions from contrast enhanced mammography (CEM) and to explore its biological mechanism.

MATERIALS AND METHODS

This retrospective study includes 1430 eligible patients who underwent CEM examination from June 2017 to July 2022 and were divided into a construction set (n=1101), an internal test set (n=196), and a pooled external test set (n=133). The AI model adopted RefineNet as a backbone network, and an attention sub-network, named convolutional block attention module (CBAM), was built upon the backbone for adaptive feature refinement. An XGBoost classifier was used to integrate the refined deep learning features with clinical characteristics to differentiate benign and malignant breast lesions. The authors further retrained the AI model to distinguish in situ and invasive carcinoma among breast cancer candidates. RNA-sequencing data from 12 patients were used to explore the underlying biological basis of the AI prediction.

RESULTS

The AI model achieved an area under the curve of 0.932 in diagnosing benign and malignant breast lesions in the pooled external test set, better than the best-performing deep learning model, radiomics model, and radiologists. Moreover, the AI model has also achieved satisfactory results (an area under the curve from 0.788 to 0.824) for the diagnosis of in situ and invasive carcinoma in the test sets. Further, the biological basis exploration revealed that the high-risk group was associated with the pathways such as extracellular matrix organization.

CONCLUSIONS

The AI model based on CEM and clinical characteristics had good predictive performance in the diagnosis of breast lesions.

Performance of mid-upper arm circumference to identify adolescents with obesity and metabolic syndrome: NHANES 2011-2018 analysis

BACKGROUND

Mid-upper arm circumference (MUAC) was recommended for screening of adolescents with obesity, although its diagnostic performance with respect to high-precision assessment of body composition remains unknown.

OBJECTIVE

To evaluate the diagnostic performance of MUAC in identifying obesity and metabolic syndrome in U.S. adolescents.

METHODS

A cross-sectional study was conducted using National Health and Nutrition Examination Survey (NHANES) data (2011-2018) of adolescents aged 12-19. We calculated the area under the receiver operating characteristic curve, sensitivity, specificity, positive and negative predictive values and likelihood ratios of MUAC in identifying obesity and metabolic syndrome.

RESULTS

In our study, data of 5496 adolescents, including 2665 females, were analysed. The prevalence of obesity was higher in boys (14%) than girls (10%), whilst metabolic syndrome was more common in males (2.6%) than females (1.7%). The area under the curve (AUC) of MUAC in identifying obesity was 0.69 in boys and 0.86 in girls, whilst the AUC of MUAC in identifying metabolic syndrome was 0.91 in boys and 0.87 in girls. The optimal MUAC cut-off for identifying adolescents with obesity was 28.3 cm in boys (sensitivity: 64.8%, specificity: 85.5%) and 30.8 cm in girls (sensitivity: 67.9%, specificity: 90.1%).

CONCLUSIONS

MUAC was a good indicator of both obesity and metabolic syndrome, with higher accuracy in girls.

Diagnostic methods for dysthyroid optic neuropathy: A systematic review and analysis

Diagnosis of dysthyroid optic neuropathy (DON) typically relies on a set of diagnostic clinical features, including decreased visual acuity, impaired color vision, presence of relative afferent pupillary defect, optic disc swelling and ancillary tests including visual field (VF), pattern visual evoked potential (pVEP), and apical crowding or optic nerve stretching on neuroimaging. We summarize various diagnostic methods to establish or rule out DON. A total of 95 studies (involving 4619 DON eyes) met the inclusion criteria. All of the studies considered clinical features as evidence of DON, while most of the studies confirmed DON diagnosis by combining clinical features with ancillary tests. Forty studies (42.1%) used at least 2 out of the 3 tests (VF, pVEP and neuroimaging) and 13 studies (13.7%) used all 3 tests to diagnose DON. In 64 % of the published studies regarding DON, the diagnostic methods of DON were not specified. It is important to note the limitations of relying solely on clinical features for diagnosing DON. On the other hand, since some eyes with optic neuropathy can be normal in one ancillary test, but abnormal in another, using more than one ancillary test to aid diagnosis is crucial and should be interpreted in correlation with clinical features. We found that the diagnostic methods of DON in most studies involved using a combination of specific clinical features and at least 2 ancillary tests.

Performance of ICD-10-AM codes for quality improvement monitoring of hospital-acquired pneumonia in a haematology-oncology casemix in Victoria, Australia

BACKGROUND

The Australian hospital-acquired complication (HAC) policy was introduced to facilitate negative funding adjustments in Australian hospitals using ICD-10-AM codes.

OBJECTIVE

The aim of this study was to determine the positive predictive value (PPV) of the ICD-10-AM codes in the HAC framework to detect hospital-acquired pneumonia in patients with cancer and to describe any change in PPV before and after implementation of an electronic medical record (EMR) at our centre.

METHOD

A retrospective case review of all coded pneumonia episodes at the Peter MacCallum Cancer Centre in Melbourne, Australia spanning two time periods (01 July 2015 to 30 June 2017 [pre-EMR period] and 01 September 2020 to 28 February 2021 [EMR period]) was performed to determine the proportion of events satisfying standardised surveillance definitions.

RESULTS

HAC-coded pneumonia occurred in 3.66% (n = 151) of 41,260 separations during the study period. Of the 151 coded pneumonia separations, 27 satisfied consensus surveillance criteria, corresponding to an overall PPV of 0.18 (95% CI: 0.12, 0.25). The PPV was approximately three times higher following EMR implementation (0.34 [95% CI: 0.19, 0.53] versus 0.13 [95% CI: 0.08, 0.21]; p = .013).

CONCLUSION

The current HAC definition is a poor-to-moderate classifier for hospital-acquired pneumonia in patients with cancer and, therefore, may not accurately reflect hospital-level quality improvement. Implementation of an EMR did enhance case detection, and future refinements to administratively coded data in support of robust monitoring frameworks should focus on EMR systems.

IMPLICATIONS

Although ICD-10-AM data are readily available in Australian healthcare settings, these data are not sufficient for monitoring and reporting of hospital-acquired pneumonia in haematology-oncology patients.

Diagnostic accuracy of clinical signs to detect erosive tooth wear in its early phase

BACKGROUND

Agreement exists about most of the clinical features of erosive tooth wear, though no evidence supports their validity in diagnosing the condition.

OBJECTIVE

This study aimed to determine the accuracy of clinical signs for diagnosing erosive tooth wear in a young adult general population.

METHODS

We conducted a cross-sectional study of dental students. In the first session, two examiners independently determined the presence of erosive tooth wear based on glazed enamel surfaces, morphological changes on non-occlusal surfaces, flattening of convex areas, or any type of concavity. In the second session, one examiner recorded the presence of clinical signs according to the Tooth Wear Evaluation System. The diagnostic accuracy of each clinical sign, both alone and combined, was assessed by calculating their sensitivity and specificity for detecting erosive tooth wear and performing multivariate logistic regression models.

RESULTS

Of the 147 participants (78 women and 69 men; median age, 22 years) we included, 76.2% had erosive tooth wear. The single clinical signs with greatest balance between the sensitivity and specificity were 'convex areas flatten' (63% and 71%, respectively) and 'dull surface' (47% and 89%, respectively). Multivariate logistic regression revealed that 'preservation of the enamel cuff' (odds ratio, 22) and the combination of 'smooth silky shining, silky glazed appearance, and dull surface' (odds ratio, 68) had the best predictive values.

CONCLUSIONS

The most accurate clinical signs for detecting early erosive tooth wear were dull surface, flattened convex areas and preservation of the enamel cuff.

Validation of Thermal Imaging and the ALT-70 Prediction Model to Differentiate Cellulitis From Pseudocellulitis

Importance

Cellulitis is misdiagnosed in up to 30% of cases due to mimic conditions termed pseudocellulitis. The resulting overuse of antibiotics is a threat to patient safety and public health. Surface thermal imaging and the ALT-70 (asymmetry, leukocytosis, tachycardia, and age ≥70 years) prediction model have been proposed as tools to help differentiate cellulitis from pseudocellulitis.

Objectives

To validate differences in skin surface temperatures between patients with cellulitis and patients with pseudocellulitis, assess the optimal temperature measure and cut point for differentiating cellulitis from pseudocellulitis, and compare the performance of skin surface temperature and the ALT-70 prediction model in differentiating cellulitis from pseudocellulitis.

Design, Setting, and Participants

This prospective diagnostic validation study was conducted among patients who presented to the emergency department with acute dermatologic lower extremity symptoms from October 11, 2018, through March 11, 2020. Statistical analysis was performed from July 2020 to March 2021 with additional work conducted in September 2023.

Main Outcomes and Measures

Temperature measures for affected and unaffected skin were obtained. Cellulitis vs pseudocellulitis was assessed by a 6-physician, independent consensus review. Differences in temperature measures were compared using the t test. Logistic regression was used to identify the temperature measure and associated cut point with the optimal performance for discriminating between cellulitis and pseudocellulitis. Diagnostic performance characteristics for the ALT-70 prediction model, surface skin temperature, and both combined were also assessed.

Results

The final sample included 204 participants (mean [SD] age, 56.6 [16.5] years; 121 men [59.3%]), 92 (45.1%) of whom had a consensus diagnosis of cellulitis. There were statistically significant differences in all skin surface temperature measures (mean temperature, maximum temperature, and gradients) between cellulitis and pseudocellulitis. The maximum temperature of the affected limb for patients with cellulitis was 33.2 °C compared with 31.2 °C for those with pseudocellulitis (difference, 2.0 °C [95% CI, 1.3-2.7 °C]; P < .001). The maximum temperature was the optimal temperature measure with a cut point of 31.2 °C in the affected skin, yielding a mean (SD) negative predictive value of 93.5% (4.7%) and a sensitivity of 96.8% (2.3%). The sensitivity of all 3 measures remained above 90%, while specificity varied considerably (ALT-70, 22.0% [95% CI, 15.8%-28.1%]; maximum temperature of the affected limb, 38.4% [95% CI, 31.7%-45.1%]; combination measure, 53.9% [95% CI, 46.5%-61.2%]).

Conclusions and Relevance

In this large diagnostic validation study, significant differences in skin surface temperature measures were observed between cases of cellulitis and cases of pseudocellulitis. Thermal imaging and the ALT-70 both demonstrated high sensitivity, but specificity was improved by combining the 2 measures. These findings support the potential of thermal imaging, alone or in combination with the ALT-70 prediction model, as a diagnostic adjunct that may reduce overdiagnosis of cellulitis.

Troponin testing in routine primary care: observations from a dynamic cohort study in the Amsterdam metropolitan area

OBJECTIVES

Troponin testing is indicated in the diagnostic work-up of acute coronary syndrome (ACS) and incorporated in risk stratification pathways. This study aims to gain insights on the use, outcomes, and diagnostic accuracy of troponin testing in routine primary care; a setting that is understudied.

METHODS

Routine data were used from the academic primary care network in the Amsterdam metropolitan area (968,433 patient records). The study population included adult patients who underwent high-sensitivity troponin I or T (hs-TnI/T) testing between 2011 and 2021. The primary outcome was the reported diagnosis and the secondary outcome was the diagnostic accuracy measured by death or ACS at 30 days.

RESULTS

3,184 patients underwent hs-troponin testing, either with hsTNT (n=2,333) or hsTNI (n=851). Median patients' age was 55 (44-65) years, and 62.3 % were female. Predominant symptoms were chest pain and dyspnea (56.7 %). Additional diagnostic laboratory tests were commonly performed (CRP: 47.7 %, natriuretic peptides: 25.6 %, d-dimer: 21.5 %). Most common diagnoses were musculoskeletal symptoms (21.6 %) and coronary heart disease (7.1 %; 1.1 % ACS). Troponin testing showed sensitivity and specificity of 77.8 % (60.9-89.9) and 94.3 % (93.5-95.1), respectively. Negative and positive predictive values were 99.7 (99.5-99.9) and 13.5 (11.1-16.4), and positive and negative likelihood ratios were 13.7 (10.9-17.1) and 0.24 (0.13-0.43).

CONCLUSIONS

GPs occasionally use troponin testing in very low-risk patients, often as part of a multi-marker rule-out strategy. The diagnostic characteristics of troponin tests, while promising, warrant prospective validation and implementation to facilitate appropriate use.

Deep learning combining mammography and ultrasound images to predict the malignancy of BI-RADS US 4A lesions in women with dense breasts: a diagnostic study

OBJECTIVES

The authors aimed to assess the performance of a deep learning (DL) model, based on a combination of ultrasound (US) and mammography (MG) images, for predicting malignancy in breast lesions categorized as Breast Imaging Reporting and Data System (BI-RADS) US 4A in diagnostic patients with dense breasts.

METHODS

A total of 992 patients were randomly allocated into the training cohort and the test cohort at a proportion of 4:1. Another, 218 patients were enrolled to form a prospective validation cohort. The DL model was developed by incorporating both US and MG images. The predictive performance of the combined DL model for malignancy was evaluated by sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The combined DL model was then compared to a clinical nomogram model and to the DL model trained using US image only and to that trained MG image only.

RESULTS

The combined DL model showed satisfactory diagnostic performance for predicting malignancy in breast lesions, with an AUC of 0.940 (95% CI: 0.874-1.000) in the test cohort, and an AUC of 0.906 (95% CI: 0.817-0.995) in the validation cohort, which was significantly higher than the clinical nomogram model, and the DL model for US or MG alone ( P <0.05).

CONCLUSIONS

The study developed an objective DL model combining both US and MG imaging features, which was proven to be more accurate for predicting malignancy in the BI-RADS US 4A breast lesions of patients with dense breasts. This model may then be used to more accurately guide clinicians' choices about whether performing biopsies in breast cancer diagnosis.

Intrinsic capacity assessment using World Health Organization Integrated Care for Older People Step 1, and the association with frailty in community dwelling older adults

AIM

This study aimed to investigate the association between intrinsic capacity (IC) and frailty in community-dwelling older adults. Specifically, we examined the utility of the World Health Organization's Integrated Care for Older People Step 1 screen for identifying frail older persons in the community.

METHODS

This is a cross-sectional analysis of a community frailty screening initiative. IC loss was ascertained using the World Health Organization's Integrated Care for Older People Step 1 questions. The Clinical Frailty Scale was used to categorize participants as robust (Clinical Frailty Scale S1-3) or frail (Clinical Frailty Scale ≥4). Logistic regression was used to analyze the association of individual and cumulative IC losses with frailty, adjusting for confounders. Additionally, the diagnostic performance of using cumulative IC losses to identify frailty was assessed.

RESULTS

This study included 1164 participants (28.2% frail). Loss in locomotion (adjusted odds ratio [AOR] 1.47, 95% CI 1.07-2.02), vitality (AOR 1.58, 95% CI 1.04-2.39), sensory (AOR 1.99, 95% CI 1.51-2.64) and psychological capacities (AOR 1.92, 95% CI 1.45-2.56) were significantly associated with frailty. Loss in more than three IC domains was associated with frailty. Using loss in at least three ICs identifies frailty, with sensitivity of 38.6%, specificity of 83.5% and positive predictive value of 47.4%. Using loss in at least four ICs improved specificity to 96.9%, and is associated with the highest positive predictive value of 57.6% and highest positive likelihood ratio of 3.55 for frailty among all cut-off values. The area under the receiver operating characteristic curve was 0.64 (95% CI 0.61-0.68).

CONCLUSIONS

IC loss as identified through World Health Organization's Integrated Care for Older People Step 1 is associated with frailty community-dwelling older adults. Geriatr Gerontol Int 2024; 24: 457-463.

"Diagnostic ability in oral pathology among different population clusters"

OBJECTIVES

To assess the diagnostic ability in detecting oral lesions among dentists, dental hygienists, dentistry students, oral hygiene students, and non-healthcare subjects.

MATERIALS AND METHODS

Participants were invited to classify 30 images of oral lesions in "benign" or "suspected malignant" only based on the visual appearance of the lesion. Diagnostic accuracy was assessed by calculating sensitivity and specificity with 95% confidence intervals and stratified by population group and image features (color, shape, and size of the lesions).

RESULTS

A total of 16,590 examinations by 553 subjects were analyzed. Overall sensitivity and specificity were 57% (95% confidence interval 56%-58%) and 64% (95% confidence interval 63%-65%). Diagnostic accuracy varied among population groups, with experienced dentists showing the lowest sensitivity (52%) and the highest specificity (71%). Red lesions, flat lesions, and large lesions had the lowest sensitivity (42%, 36%, 57%) but the highest specificity (70%, 75%, 76%).

CONCLUSIONS

We found worrying low ability to detect suspected malignant oral lesions by both healthcare workers and non-healthcare subjects. Lesion-specific characteristics may lead to differences in recognition. Specific courses and more adequate teaching methods should be proposed to increase identification of oral lesions.

Dynamic changes in serum (1-3)-β-D-glucan caused by intravenous immunoglobulin infusion: A prospective study

PURPOSE

The purpose of this study was to investigate the dynamic changes in serum (1-3)-β-D-glucan (BDG) caused by intravenous immunoglobulins (IVIG) infusion in adults.

METHODS

This study included patients who received IVIG infusion from October 2021 to October 2022 during hospitalization. We randomly examined two IVIG samples for every patient. Serum samples were collected at nine time points: before (Tpre), immediately (T1-0), 6h (T1-1) and 12h (T1-2) later on the first day; immediately (T2-0) and six hours later (T2-1) on the second day during IVIG infusion, and within three days after IVIG infusion (Ta1, Ta2, and Ta3, respectively). The Friedman test was used for statistical analysis.

RESULTS

A total of 159 serum BDG from 19 patients were included in the analysis. The BDG content of IVIG ranged from 249 pg/ml to 4812 pg/ml. Patients had significantly elevated serum BDG on T1-0 (176 (113, 291) pg/ml, p = 0.002) and Ta1 (310 (199, 470) pg/ml, p < 0.001), compared with Tpre (41 (38, 65) pg/ml). The increments of serum BDG (ΔBDG) were associated with BDG concentration of IVIG (Spearman r = 0.59, p = 0.02). Individuals with abnormal renal function indexes showed higher serum ΔBDG values at Ta1 (403 (207, 484) pg/ml) than patients with normal renal function (172 (85, 316) pg/ml, p = 0.036).

CONCLUSION

Patients who received IVIG had significantly higher serum BDG values. Elevated BDG levels correlate with BDG content of IVIG and abnormal renal function indexes.

In-Laboratory Polysomnography Worsens Obstructive Sleep Apnea by Changing Body Position Compared to Home Testing

(1) Background: Home sleep apnea testing, known as polysomnography type 3 (PSG3), underestimates respiratory events in comparison with in-laboratory polysomnography type 1 (PSG1). Without head electrodes for scoring sleep and arousal, in a home environment, patients feel unfettered and move their bodies more naturally. Adopting a natural position may decrease obstructive sleep apnea (OSA) severity in PSG3, independently of missing hypopneas associated with arousals. (2) Methods: Patients with suspected OSA performed PSG1 and PSG3 in a randomized sequence. We performed an additional analysis, called reduced polysomnography, in which we blindly reassessed all PSG1 tests to remove electroencephalographic electrodes, electrooculogram, and surface electromyography data to estimate the impact of not scoring sleep and arousal-based hypopneas on the test results. A difference of 15 or more in the apnea-hypopnea index (AHI) between tests was deemed clinically relevant. We compared the group of patients with and without clinically relevant differences between lab and home tests (3) Results: As expected, by not scoring sleep, there was a decrease in OSA severity in the lab test, similar to the home test results. The group of patients with clinically relevant differences between lab and home tests presented more severe OSA in the lab compared to the other group (mean AHI, 42.5 vs. 20.2 events/h, p = 0.002), and this difference disappeared in the home test. There was no difference between groups in the shift of OSA severity by abolishing sleep scoring in the lab. However, by comparing lab and home tests, there were greater variations in supine AHI and time spent in the supine position in the group with a clinically relevant difference, either with or without scoring sleep, showing an impact of the site of the test on body position during sleep. These variations presented as a marked increase or decrease in supine outcomes according to the site of the test, with no particular trend. (4) Conclusions: In-lab polysomnography may artificially increase OSA severity in a subset of patients by inducing marked changes in body position compared to home tests. The location of the sleep test seems to interfere with the evaluation of patients with more severe OSA.

Role of Inflammatory Markers in the Assessment of Meningitis in Adult Patients with Fever and Headache

BACKGROUND

Meningitis, especially of bacterial origin, is a medical emergency that must be diagnosed promptly. However, due to the associated risks of complications of lumbar puncture, it is crucial to identify individuals who truly need it. The aim of this study was to assess the diagnostic role of inflammatory markers in distinguishing among patients without meningitis, those with aseptic meningitis, and those with bacterial meningitis.

METHODS

This was a retrospective, diagnostic study at an acute care hospital, involving adult patients who presented to either ambulatory care or the emergency department with fever and headache, but without altered mental status or neurological deficits. Inflammatory markers (C-reactive protein [CRP], mean platelet volume, neutrophil-lymphocyte ratio, and red cell distribution width) were assessed as index tests. An expert panel classified patients into three groups: no meningitis, aseptic meningitis, and bacterial meningitis using predefined criteria.

RESULTS

Of the 80 patients, 52 had no meningitis, 27 had aseptic meningitis, and 1 had bacterial meningitis. Of the inflammatory markers investigated, only CRP showed potential usefulness in differentiating these three diagnostic groups, with median values of 5.6 (interquartile range [IQR] 2.1, 11.3) mg/dL in those without meningitis, 0.2 (IQR 0.1, 1.2) mg/dL in those with aseptic meningitis, and notably elevated at 21.7 mg/dL in the patient with bacterial meningitis.

CONCLUSION

In adult patients presenting with fever and headache in an emergency setting, CRP was the only marker that demonstrated potential diagnostic utility in distinguishing among those with no meningitis, aseptic meningitis, and bacterial meningitis.

Home-Monitoring Vision Tests to Detect Active Neovascular Age-Related Macular Degeneration

Importance

Most neovascular age-related macular degeneration (nAMD) treatments involve long-term follow-up of disease activity. Home-monitoring would reduce the burden on patients and their caregivers and release clinic capacity.

Objective

To evaluate 3 vision home-monitoring tests for patients to use to detect active nAMD compared with diagnosing active nAMD at hospital follow-up during the after-treatment monitoring phase.

Design, Setting, and Participants

This was a diagnostic test accuracy study wherein the reference standard was detection of active nAMD by an ophthalmologist at hospital follow-up. The 3 home-monitoring tests evaluated included the following: (1) the KeepSight Journal (KSJ [International Macular and Retinal Foundation]), which contains paper-based near-vision tests presented as word puzzles, (2) the MyVisionTrack (mVT [Genentech]) vision-monitoring mobile app, viewed on an Apple mobile operating system-based device, and (3) the MultiBit (MBT [Visumetrics]) app, viewed on an Apple mobile operating system-based device. Participants were asked to test weekly; mVT and MBT scores were transmitted automatically, and KSJ scores were returned to the research office every 6 months. Raw scores between hospital follow-ups were summarized as averages. Patients were recruited from 6 UK hospital eye clinics and were 50 years and older with at least 1 eye first treated for active nAMD for at least 6 months or longer to a maximum of 42 months before approach. Participants were stratified by time since starting treatment. Study data were analyzed from May to September 2021.

Exposures

The KSJ, mVT, and MBT were compared with the reference standard (in-hospital ophthalmologist examination).

Main Outcomes and Measures

Estimated area under receiver operating characteristic curve (AUROC). The study had 90% power to detect a difference of 0.06, or 80% power to detect a difference of 0.05, if the AUROC for 2 tests was 0.75.

Results

A total of 297 patients (mean [SD] age, 74.9 [6.6] years; 174 female [58.6%]) were included in the study. At least 1 hospital follow-up was available for 312 study eyes in 259 participants (1549 complete visits). Median (IQR) home-monitoring testing frequency was 3 (1-4) times per month. Estimated AUROC was less than 0.6 for all home-monitoring tests, and only the KSJ summary score was associated with lesion activity (odds ratio, 3.48; 95% CI, 1.09-11.13; P = .04).

Conclusions and Relevance

Results suggest that no home-monitoring vision test evaluated provided satisfactory diagnostic accuracy to identify active nAMD diagnosed in hospital eye service follow-up clinics. Implementing any of these evaluated tests, with ophthalmologists only reviewing test positives, would mean most active lesions were missed, risking unnecessary sight loss.

Validation of self-applied unattended polysomnography using Somte V2 PSG (Somte) for diagnosis of obstructive sleep apnoea (OSA) in pregnant women in early to mid-gestation

PURPOSE

Polysomnography (PSG) may be completed in the home environment (unattended), and when self-applied, allow the collection of data with minimal healthcare worker intervention. Self-applied, unattended PSG in the home environment using Somte PSG V2 (Somte) has not been validated in pregnant women in early to mid-gestation. We undertook a study to evaluate the accuracy of Somte compared to attended PSG. The agreement between apnoea hypopnea index (AHI) and respiratory disturbance index (RDI) scores in Somte and PSG in early to mid-gestation were assessed.

METHODS

Pregnant women (≤ 24 weeks gestation) were scheduled for PSG and Somte within a 7-day window, in any order. Somte were self-applied and completed in the home. Somte were scored blinded to PSG result. AHI was the primary outcome of interest, though an AHI ≥ 5 or RDI ≥ 5 on PSG was considered diagnostic of Obstructive Sleep Apnoea (OSA). AHI, RDI, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) was calculated and receiver operating characteristic (ROC) curves were produced. Bland-Altman plots were used to determine agreement. Technical issues occurring during tests were explored.

RESULTS

Twenty-four participants successfully completed both tests between March 2021 and January 2023. PSG were completed at around 14.1 weeks' gestation (IQR 13.4, 15.7). The time interval between Somte and PSG was a median of 4 days (IQR 2, 7 (range 1-12)). Five (20.8%) women had OSA on PSG at AHI ≥ 5 and 10 (41.6%) women had OSA on PSG at RDI ≥ 5. Somte and PSG did not differ in the measurement of AHI ((1.8, 1.6, p = 0.09) or RDI (3.3, 3.5), p = 0.73). At AHI ≥ 5, diagnostic test accuracy (area under the ROC curve) of Somte was 0.94, sensitivity 80.0%, specificity 94.7%, PPV and NPV were 80.0% and 94.7% respectively. At RDI ≥ 5, diagnostic test accuracy (area under the ROC curve) was 0.95, sensitivity 60.0%, specificity 93.0% and PPV and NPV were 85.7% and 76.4% respectively. The confidence limits of Bland-Altman plots were 6.37 to - 8.89 at cut off AHI ≥ 5 and 8.89 to - 10.43 at cut off RDI ≥ 5. Somte failed to start in four tests. Technical issues were reported in both Somte (n = 13, 54.2%) and PSG (n = 6, 25.0%).

CONCLUSION

Self-applied, unattended Somte may provide an acceptable substitute to attended PSG in the identification of OSA in pregnant women in early to mid-gestation in this small sample but may fail to detect cases of OSA, particularly when using RDI as the diagnostic marker.

Detecting white spot lesions on post-orthodontic oral photographs using deep learning based on the YOLOv5x algorithm: a pilot study

BACKGROUND

Deep learning model trained on a large image dataset, can be used to detect and discriminate targets with similar but not identical appearances. The aim of this study is to evaluate the post-training performance of the CNN-based YOLOv5x algorithm in the detection of white spot lesions in post-orthodontic oral photographs using the limited data available and to make a preliminary study for fully automated models that can be clinically integrated in the future.

METHODS

A total of 435 images in JPG format were uploaded into the CranioCatch labeling software and labeled white spot lesions. The labeled images were resized to 640 × 320 while maintaining their aspect ratio before model training. The labeled images were randomly divided into three groups (Training:349 images (1589 labels), Validation:43 images (181 labels), Test:43 images (215 labels)). YOLOv5x algorithm was used to perform deep learning. The segmentation performance of the tested model was visualized and analyzed using ROC analysis and a confusion matrix. True Positive (TP), False Positive (FP), and False Negative (FN) values were determined.

RESULTS

Among the test group images, there were 133 TPs, 36 FPs, and 82 FNs. The model's performance metrics include precision, recall, and F1 score values of detecting white spot lesions were 0.786, 0.618, and 0.692. The AUC value obtained from the ROC analysis was 0.712. The mAP value obtained from the Precision-Recall curve graph was 0.425.

CONCLUSIONS

The model's accuracy and sensitivity in detecting white spot lesions remained lower than expected for practical application, but is a promising and acceptable detection rate compared to previous study. The current study provides a preliminary insight to further improved by increasing the dataset for training, and applying modifications to the deep learning algorithm.

CLINICAL REVELANCE

Deep learning systems can help clinicians to distinguish white spot lesions that may be missed during visual inspection.

C-reactive protein-based tuberculosis triage testing: a multi-country diagnostic accuracy study

Rationale

C-reactive protein (CRP)-based tuberculosis (TB) screening is recommended for people with HIV (PWH). However, its performance among people without HIV and in diverse settings is unknown.

Objectives

In a multi-country study, we aimed to determine whether CRP meets the minimum accuracy targets (sensitivity ≥90%, specificity ≥70%) for an effective TB triage test.

Methods/Measurements

Consecutive outpatient adults with cough ≥2 weeks from five TB endemic countries in Africa and Asia had baseline blood collected for point-of-care CRP testing and HIV and diabetes screening. Sputum samples were collected for Xpert MTB/RIF Ultra (Xpert) testing and culture. CRP sensitivity and specificity (5 mg/L cut-point) was determined in reference to sputum test results and compared by country, sex, and HIV and diabetes status. Variables affecting CRP performance were identified using a multivariate receiver operating characteristic (ROC) regression model.

Results

Among 2904 participants, of whom 613 (21%) had microbiologically-confirmed TB, CRP sensitivity was 84% (95% CI: 81-87%) and specificity was 61% (95% CI: 59-63%). CRP accuracy varied geographically, with higher sensitivity in African countries (≥91%) than Asian countries (64-82%). Sensitivity was higher among men than women (87% vs. 79%, difference +8%, 95% CI: 1-15%) and specificity was higher among people without HIV than PWH (64% vs. 45%, difference +19%, 95% CI: 13-25%). ROC regression identified country and measures of TB disease severity as predictors of CRP performance.

Conclusions

Overall, CRP did not achieve the minimum accuracy targets and its performance varied by setting and in some sub-groups, likely reflecting population differences in mycobacterial load.

Wide Real-Life Data Support Reduced Sensitivity of Antigen Tests for Omicron SARS-CoV-2 Infections

With the continuous spread of new SARS-CoV-2 variants of concern (VOCs), the monitoring of diagnostic test performances is mandatory. We evaluated the changes in antigen diagnostic tests' (ADTs) accuracy along the Delta to Omicron VOCs transition, exploring the N protein mutations possibly affecting ADT sensitivity and assessing the best sampling site for the diagnosis of Omicron infections. In total, 5175 subjects were enrolled from 1 October 2021 to 15 July 2022. The inclusion criteria were SARS-CoV-2 ADT combined with a same-day RT-PCR swab test. For the sampling site analysis, 61 patients were prospectively recruited during the Omicron period for nasal and oral swab analyses by RT-PCR. Next-Generation Sequencing data were obtained to evaluate the different sublineages. Using RT-PCR as a reference, 387 subjects resulted in becoming infected and the overall sensitivity of the ADT decreased from 63% in the Delta period to 33% in the Omicron period. This decrease was highly statistically significant (p < 0.001), and no decrease in viral load was detected at the RNA level. The nasal site presented a significantly higher viral load than the oral site during the Omicron wave. The reduced detection rate of Omicron infections by ADT should be considered in the global testing strategy to preserve accurate diagnoses across the changing SARS-CoV-2 variants.

Comparison of the Sensitivity and Specificity of Commercial Anti-Dengue Virus IgG Tests to Identify Persons Eligible for Dengue Vaccination

The Advisory Committee on Immunization Practices (ACIP) recommended that dengue pre-vaccination screening tests for Dengvaxia administration have at least 98% specificity and 75% sensitivity. This study evaluates the performance of commercial anti-DENV IgG tests to identify tests that could be used for pre-vaccination screening. First, for 7 tests, we evaluated sensitivity and specificity in early convalescent dengue virus (DENV) infection, using 44 samples collected 7-30 days after symptom onset and confirmed by RT-PCR. Next, for the 5 best performing tests and two additional tests (with and without an external test reader) that became available later, we evaluated performance to detect past dengue infection among a panel of 44 specimens collected in 2018-2019 from healthy 9-16-year-old children from Puerto Rico. Finally, a full-scale evaluation was done with the 4 best performing tests using 400 specimens from the same population. We used virus focus reduction neutralization test and an in-house DENV IgG ELISA as reference standards. Of seven tests, five showed ≥75% sensitivity detecting anti-DENV IgG in early convalescent specimens with low cross-reactivity to Zika virus. For the detection of previous DENV infections the tests with the highest performance were the Euroimmun NS1 IgG ELISA (sensitivity 84.5%, specificity 97.1%) and CTK Dengue IgG rapid test R0065C with the test reader (sensitivity 76.2% specificity 98.1%). There are IgG tests available that can be used to accurately classify individuals with previous DENV infection as eligible for dengue vaccination to support safe vaccine implementation.