Evaluation of a decided sample size in machine learning applications

A guide to artificial intelligence for cancer researchers

Artificial intelligence (AI) has been commoditized. It has evolved from a specialty resource to a readily accessible tool for cancer researchers. AI-based tools can boost research productivity in daily workflows, but can also extract hidden information from existing data, thereby enabling new scientific discoveries. Building a basic literacy in these tools is useful for every cancer researcher. Researchers with a traditional biological science focus can use AI-based tools through off-the-shelf software, whereas those who are more computationally inclined can develop their own AI-based software pipelines. In this article, we provide a practical guide for non-computational cancer researchers to understand how AI-based tools can benefit them. We convey general principles of AI for applications in image analysis, natural language processing and drug discovery. In addition, we give examples of how non-computational researchers can get started on the journey to productively use AI in their own work.

Challenges and Potential of Artificial Intelligence in Neuroradiology

PURPOSE

Artificial intelligence (AI) has emerged as a transformative force in medical research and is garnering increased attention in the public consciousness. This represents a critical time period in which medical researchers, healthcare providers, insurers, regulatory agencies, and patients are all developing and shaping their beliefs and policies regarding the use of AI in the healthcare sector. The successful deployment of AI will require support from all these groups. This commentary proposes that widespread support for medical AI must be driven by clear and transparent scientific reporting, beginning at the earliest stages of scientific research.

METHODS

A review of relevant guidelines and literature describing how scientific reporting plays a central role at key stages in the life cycle of an AI software product was conducted. To contextualize this principle within a specific medical domain, we discuss the current state of predictive tissue outcome modeling in acute ischemic stroke and the unique challenges presented therein.

RESULTS AND CONCLUSION

Translating AI methods from the research to the clinical domain is complicated by challenges related to model design and validation studies, medical product regulations, and healthcare providers' reservations regarding AI's efficacy and affordability. However, each of these limitations is also an opportunity for high-impact research that will help to accelerate the clinical adoption of state-of-the-art medical AI. In all cases, establishing and adhering to appropriate reporting standards is an important responsibility that is shared by all of the parties involved in the life cycle of a prospective AI software product.

Best of Both Worlds: Bridging One Model for All and Group-Specific Model Approaches using Ensemble-based Subpopulation Modeling

Subpopulation models have become of increasing interest in prediction of clinical outcomes because they promise to perform better for underrepresented patient subgroups. However, the personalization benefits gained from these models tradeoff their statistical power, and can be impractical when the subpopulation's sample size is small. We hypothesize that a hierarchical model in which population information is integrated into subpopulation models would preserve the personalization benefits and offset the loss of power. In this work, we integrate ideas from ensemble modeling, personalization, and hierarchical modeling and build ensemble-based subpopulation models in which specialization relies on whole group samples. This approach significantly improves the precision of the positive class, especially for the underrepresented subgroups, with minimal cost to the recall. It consistently outperforms one model for all and one model for each subgroup approaches, especially in the presence of a high class-imbalance, for subgroups with at least 380 training samples.

Interpretable machine learning decodes soil microbiome's response to drought stress

BACKGROUND

Extreme weather events induced by climate change, particularly droughts, have detrimental consequences for crop yields and food security. Concurrently, these conditions provoke substantial changes in the soil bacterial microbiota and affect plant health. Early recognition of soil affected by drought enables farmers to implement appropriate agricultural management practices. In this context, interpretable machine learning holds immense potential for drought stress classification of soil based on marker taxa.

RESULTS

This study demonstrates that the 16S rRNA-based metagenomic approach of Differential Abundance Analysis methods and machine learning-based Shapley Additive Explanation values provide similar information. They exhibit their potential as complementary approaches for identifying marker taxa and investigating their enrichment or depletion under drought stress in grass lineages. Additionally, the Random Forest Classifier trained on a diverse range of relative abundance data from the soil bacterial micobiome of various plant species achieves a high accuracy of 92.3 % at the genus rank for drought stress prediction. It demonstrates its generalization capacity for the lineages tested.

CONCLUSIONS

In the detection of drought stress in soil bacterial microbiota, this study emphasizes the potential of an optimized and generalized location-based ML classifier. By identifying marker taxa, this approach holds promising implications for microbe-assisted plant breeding programs and contributes to the development of sustainable agriculture practices. These findings are crucial for preserving global food security in the face of climate change.

Radiomics of Intrahepatic Cholangiocarcinoma and Peritumoral Tissue Predicts Postoperative Survival: Development of a CT-Based Clinical-Radiomic Model

BACKGROUND

For many tumors, radiomics provided a relevant prognostic contribution. This study tested whether the computed tomography (CT)-based textural features of intrahepatic cholangiocarcinoma (ICC) and peritumoral tissue improve the prediction of survival after resection compared with the standard clinical indices.

METHODS

All consecutive patients affected by ICC who underwent hepatectomy at six high-volume centers (2009-2019) were considered for the study. The arterial and portal phases of CT performed fewer than 60 days before surgery were analyzed. A manual segmentation of the tumor was performed (Tumor-VOI). A 5-mm volume expansion then was applied to identify the peritumoral tissue (Margin-VOI).

RESULTS

The study enrolled 215 patients. After a median follow-up period of 28 months, the overall survival (OS) rate was 57.0%, and the progression-free survival (PFS) rate was 34.9% at 3 years. The clinical predictive model of OS had a C-index of 0.681. The addition of radiomic features led to a progressive improvement of performances (C-index of 0.71, including the portal Tumor-VOI, C-index of 0.752 including the portal Tumor- and Margin-VOI, C-index of 0.764, including all VOIs of the portal and arterial phases). The latter model combined clinical variables (CA19-9 and tumor pattern), tumor indices (density, homogeneity), margin data (kurtosis, compacity, shape), and GLRLM indices. The model had performance equivalent to that of the postoperative clinical model including the pathology data (C-index of 0.765). The same results were observed for PFS.

CONCLUSIONS

The radiomics of ICC and peritumoral tissue extracted from preoperative CT improves the prediction of survival. Both the portal and arterial phases should be considered. Radiomic and clinical data are complementary and achieve a preoperative estimation of prognosis equivalent to that achieved in the postoperative setting.

Understanding older people's voice interactions with smart voice assistants: a new modified rule-based natural language processing model with human input

The COVID-19 pandemic has expedited the integration of Smart Voice Assistants (SVA) among older people. The qualitative data derived from user commands on SVA is pivotal for elucidating the engagement patterns of older individuals with such systems. However, the sheer volume of user-generated voice interaction data presents a formidable challenge for manual coding. Compounding this issue, age-related cognitive decline and alterations in speech patterns further complicate the interpretation of older users' SVA voice interactions. Conventional dictionary-based textual analysis tools, which count word frequencies, are inadequate in capturing the evolving and communicative essence of these interactions that unfold over a series of dialogues and modify with time. To address these challenges, our study introduces a novel, modified rule-based Natural Language Processing (MR-NLP) model augmented with human input. This reproducible approach capitalizes on human-derived insights to establish a lexicon of critical keywords and to formulate rules for the iterative refinement of the NLP model. English speakers, aged 50 or older and residing alone, were enlisted to engage with Amazon Alexa™ via predefined daily routines for a minimum of 30 min daily spanning three months (N = 35, mean age = 77). We amassed time-stamped, textual data comprising participants' user commands and responses from Alexa™. Initially, a subset constituting 20% of the data (1,020 instances) underwent manual coding by human coder, predicated on keywords and commands. Separately, a rule-based Natural Language Processing (NLP) methodology was employed to code the identical subset. Discrepancies arising between human coder and the NLP model programmer were deliberated upon and reconciled to refine the rule-based NLP coding framework for the entire dataset. The modified rule-based NLP approach demonstrated notable enhancements in efficiency and scalability and reduced susceptibility to inadvertent errors in comparison to manual coding. Furthermore, human input was instrumental in augmenting the NLP model, yielding insights germane to the aging adult demographic, such as recurring speech patterns or ambiguities. By disseminating this innovative software solution to the scientific community, we endeavor to advance research and innovation in NLP model formulation, subsequently contributing to the understanding of older people's interactions with SVA and other AI-powered systems.

A comprehensive analysis of key factors' impact on environmental performance: Evidence from Globe by novel super learner algorithm

This study aims to analyze comprehensively the impact of different economic and demographic factors, which affect economic development, on environmental performance. In this context, the study considers the Environmental Performance Index as the response variable, uses GDP per capita, tariff rate, tax burden, government expenditure, inflation, unemployment, population, income tax rate, public debt, FDI inflow, and corporate tax rate as the explanatory variables, examines 181 countries, performs a novel Super Learner (SL) algorithm, which includes a total of six machine learning (ML) algorithms, and uses data for the years 2018, 2020, and 2022. The results demonstrate that (i) the SL algorithm has a superior capacity with regard to other ML algorithms; (ii) gross domestic product per capita is the most crucial factor in the environmental performance followed by tariff rates, tax burden, government expenditure, and inflation, in order; (iii) among all, the corporate tax rate has the lowest importance on the environmental performance followed by also foreign direct investment, public debt, income tax rate, population, and unemployment; (iv) there are some critical thresholds, which imply that the impact of the factors on the environmental performance change according to these barriers. Overall, the study reveals the nonlinear impact of the variables on environmental performance as well as their relative importance and critical threshold. Thus, the study provides policymakers valuable insights in re-formulating their environmental policies to increase environmental performance. Accordingly, various policy options are discussed.

Association of Visual-Based Signals with Electroencephalography Patterns in Enhancing the Drowsiness Detection in Drivers with Obstructive Sleep Apnea

Individuals with obstructive sleep apnea (OSA) face increased accident risks due to excessive daytime sleepiness. PERCLOS, a recognized drowsiness detection method, encounters challenges from image quality, eyewear interference, and lighting variations, impacting its performance, and requiring validation through physiological signals. We propose visual-based scoring using adaptive thresholding for eye aspect ratio with OpenCV for face detection and Dlib for eye detection from video recordings. This technique identified 453 drowsiness (PERCLOS ≥ 0.3 || CLOSDUR ≥ 2 s) and 474 wakefulness episodes (PERCLOS < 0.3 and CLOSDUR < 2 s) among fifty OSA drivers in a 50 min driving simulation while wearing six-channel EEG electrodes. Applying discrete wavelet transform, we derived ten EEG features, correlated them with visual-based episodes using various criteria, and assessed the sensitivity of brain regions and individual EEG channels. Among these features, theta-alpha-ratio exhibited robust mapping (94.7%) with visual-based scoring, followed by delta-alpha-ratio (87.2%) and delta-theta-ratio (86.7%). Frontal area (86.4%) and channel F4 (75.4%) aligned most episodes with theta-alpha-ratio, while frontal, and occipital regions, particularly channels F4 and O2, displayed superior alignment across multiple features. Adding frontal or occipital channels could correlate all episodes with EEG patterns, reducing hardware needs. Our work could potentially enhance real-time drowsiness detection reliability and assess fitness to drive in OSA drivers.

Machine Learning Prediction of Tongue Pressure in Elderly Patients with Head and Neck Tumor: A Cross-Sectional Study

Background: This investigation sought to cross validate the predictors of tongue pressure recovery in elderly patients' post-treatment for head and neck tumors, leveraging advanced machine learning techniques. Methods: By employing logistic regression, support vector regression, random forest, and extreme gradient boosting, the study analyzed an array of variables including patient demographics, surgery types, dental health status, and age, drawn from comprehensive medical records and direct tongue pressure assessments. Results: Among the models, logistic regression emerged as the most effective, demonstrating an accuracy of 0.630 [95% confidence interval (CI): 0.370-0.778], F1 score of 0.688 [95% confidence interval (CI): 0.435-0.853], precision of 0.611 [95% confidence interval (CI): 0.313-0.801], recall of 0.786 [95% confidence interval (CI): 0.413-0.938] and an area under the receiver operating characteristic curve of 0.626 [95% confidence interval (CI): 0.409-0.806]. This model distinctly highlighted the significance of glossectomy (p = 0.039), the presence of functional teeth (p = 0.043), and the patient's age (p = 0.044) as pivotal factors influencing tongue pressure, setting the threshold for statistical significance at p < 0.05. Conclusions: The analysis underscored the critical role of glossectomy, the presence of functional natural teeth, and age as determinants of tongue pressure in logistics regression, with the presence of natural teeth and the tumor site located in the tongue consistently emerging as the key predictors across all computational models employed in this study.

Weight gained during treatment predicts 6-month body mass index in a large sample of patients with anorexia nervosa using ensemble machine learning

OBJECTIVE

This study used machine learning methods to analyze data on treatment outcomes from individuals with anorexia nervosa admitted to a specialized eating disorders treatment program.

METHODS

Of 368 individuals with anorexia nervosa (209 adolescents and 159 adults), 160 individuals had data available for a 6-month follow-up analysis. Participants were treated in a 6-day-per-week partial-hospital program. Participants were assessed for eating disorder-specific and non-specific psychopathology. The analyses used established machine learning procedures combined in an ensemble model from support vector machine learning, random forest prediction, and the elastic net regularized regression with an exploration (training; 75%) and confirmation (test; 25%) split of the data.

RESULTS

The models predicting body mass index (BMI) at 6-month follow-up explained a 28.6% variance in the training set (n = 120). The model had good performance in predicting 6-month BMI in the test dataset (n = 40), with predicted BMI significantly correlating with actual BMI (r = .51, p = 0.01). The change in BMI from admission to discharge was the most important predictor, strongly correlating with reported BMI at 6-month follow-up (r = .55). Behavioral variables were much less predictive of BMI outcome. Results were similar for z-transformed BMI in the adolescent-only group. Length of stay was most predictive of weight gain in treatment (r = .56) but did not predict longer-term BMI.

CONCLUSIONS

This study, using an agnostic ensemble machine learning approach in the largest to-date sample of individuals with anorexia nervosa, suggests that achieving weight gain goals in treatment predicts longer-term weight-related outcomes. Other potential predictors, personality, mood, or eating disorder-specific symptoms were relatively much less predictive.

PUBLIC SIGNIFICANCE

The results from this study indicate that the amount of weight gained during treatment predicts BMI 6 months after discharge from a high level of care. This suggests that patients require sufficient time in a higher level of care treatment to meet their specific weight goals and be able to maintain normal weight.

Artificial Intelligence Assisted Recognition of Anatomical Landmarks and Laparoscopic Instruments in Transabdominal Preperitoneal Inguinal Hernia Repair

Laparoscopic TAPP (Trans-Abdominal PrePeritoneal) is a minimally invasive surgical procedure used to repair inguinal hernias. Arguably, one important aspect to TAPP hernia repair is the identification of anatomical landmarks and the correct use of various laparoscopic instruments. There are very few studies regarding the use of artificial intelligence in laparoscopic inguinal hernia repair and more specifically in TAPP. The aim of this study is to evaluate the feasibility and usefulness of AI in the recognition of anatomical landmarks and tools in laparoscopic TAPP videos. Imaging data have been exported from 20 Laparoscopic TAPP videos that have been performed by the authors and another 5 high quality TAPP videos from the internet (free access) performed by other surgeons. In total 1095 selected images have been exported for annotation. To accomplish the AI result of computer vision, the YOLOv8 model of deep learning was used. In total 2716 segmented areas of interest have been exported. The AI model was able to detect the various classes with a maximum F1 score of .82 when the confidence threshold was set to .406. The mAP50 was .873 for all classes. The Precision was above 50% when the confidence was over 10%. The Recall rate was above 50% when confidence was less than 70%. These results suggest that the model is effective at balancing precision and recall, capturing both true positives and minimizing false negatives. Artificial Intelligence recognition of anatomical landmarks and laparoscopic instruments in TAPP is feasible with acceptable success rates.

Development of short forms for screening children's dental caries and urgent treatment needs using item response theory and machine learning methods

OBJECTIVES

Surveys can assist in screening oral diseases in populations to enhance the early detection of disease and intervention strategies for children in need. This paper aims to develop short forms of child-report and proxy-report survey screening instruments for active dental caries and urgent treatment needs in school-age children.

METHODS

This cross-sectional study recruited 497 distinct dyads of children aged 8-17 and their parents between 2015 to 2019 from 14 dental clinics and private practices in Los Angeles County. We evaluated responses to 88 child-reported and 64 proxy-reported oral health questions to select and calibrate short forms using Item Response Theory. Seven classical Machine Learning algorithms were employed to predict children's active caries and urgent treatment needs using the short forms together with family demographic variables. The candidate algorithms include CatBoost, Logistic Regression, K-Nearest Neighbors (KNN), Naïve Bayes, Neural Network, Random Forest, and Support Vector Machine. Predictive performance was assessed using repeated 5-fold nested cross-validations.

RESULTS

We developed and calibrated four ten-item short forms. Naïve Bayes outperformed other algorithms with the highest median of cross-validated area under the ROC curve. The means of best testing sensitivities and specificities using both child-reported and proxy-reported responses were 0.84 and 0.30 for active caries, and 0.81 and 0.31 for urgent treatment needs respectively. Models incorporating both response types showed a slightly higher predictive accuracy than those relying on either child-reported or proxy-reported responses.

CONCLUSIONS

The combination of Item Response Theory and Machine Learning algorithms yielded potentially useful screening instruments for both active caries and urgent treatment needs of children. The survey screening approach is relatively cost-effective and convenient when dealing with oral health assessment in large populations. Future studies are needed to further leverage the customize and refine the instruments based on the estimated item characteristics for specific subgroups of the populations to enhance predictive accuracy.

Incorporating reproductive system history data into cardiovascular nursing research to advance women's health

The lack of sex-specific variables, such as reproductive system history (RSH), in cardiovascular research studies is a missed opportunity to address the cardiovascular disease (CVD) burden, especially among women who face sex-specific risks of developing CVD. Collecting RSH data from women enrolled in research studies is an important step towards improving women's cardiovascular health. In this paper, we describe two approaches to collecting RSH in CVD research: extracting RSH from the medical record and participant self-report of RSH. We provide specific examples from our own research and address common data management and statistical analysis problems when dealing with RSH data in research.

Variational mode decomposition-based EEG analysis for the classification of disorders of consciousness

Aberrant alterations in any of the two dimensions of consciousness, namely awareness and arousal, can lead to the emergence of disorders of consciousness (DOC). The development of DOC may arise from more severe or targeted lesions in the brain, resulting in widespread functional abnormalities. However, when it comes to classifying patients with disorders of consciousness, particularly utilizing resting-state electroencephalogram (EEG) signals through machine learning methods, several challenges surface. The non-stationarity and intricacy of EEG data present obstacles in understanding neuronal activities and achieving precise classification. To address these challenges, this study proposes variational mode decomposition (VMD) of EEG before feature extraction along with machine learning models. By decomposing preprocessed EEG signals into specified modes using VMD, features such as sample entropy, spectral entropy, kurtosis, and skewness are extracted across these modes. The study compares the performance of the features extracted from VMD-based approach with the frequency band-based approach and also the approach with features extracted from raw-EEG. The classification process involves binary classification between unresponsive wakefulness syndrome (UWS) and the minimally conscious state (MCS), as well as multi-class classification (coma vs. UWS vs. MCS). Kruskal-Wallis test was applied to determine the statistical significance of the features and features with a significance of p < 0.05 were chosen for a second round of classification experiments. Results indicate that the VMD-based features outperform the features of other two approaches, with the ensemble bagged tree (EBT) achieving the highest accuracy of 80.5% for multi-class classification (the best in the literature) and 86.7% for binary classification. This approach underscores the potential of integrating advanced signal processing techniques and machine learning in improving the classification of patients with disorders of consciousness, thereby enhancing patient care and facilitating informed treatment decision-making.

Deep Learning Enables Spatial Mapping of the Mosaic Microenvironment of Myeloma Bone Marrow Trephine Biopsies

Bone marrow trephine biopsy is crucial for the diagnosis of multiple myeloma. However, the complexity of bone marrow cellular, morphologic, and spatial architecture preserved in trephine samples hinders comprehensive evaluation. To dissect the diverse cellular communities and mosaic tissue habitats, we developed a superpixel-inspired deep learning method (MoSaicNet) that adapts to complex tissue architectures and a cell imbalance aware deep learning pipeline (AwareNet) to enable accurate detection and classification of rare cell types in multiplex immunohistochemistry images. MoSaicNet and AwareNet achieved an AUC of >0.98 for tissue and cellular classification on separate test datasets. Application of MoSaicNet and AwareNet enabled investigation of bone heterogeneity and thickness as well as spatial histology analysis of bone marrow trephine samples from monoclonal gammopathies of undetermined significance (MGUS) and from paired newly diagnosed and posttreatment multiple myeloma. The most significant difference between MGUS and newly diagnosed multiple myeloma (NDMM) samples was not related to cell density but to spatial heterogeneity, with reduced spatial proximity of BLIMP1+ tumor cells to CD8+ cells in MGUS compared with NDMM samples. Following treatment of patients with multiple myeloma, there was a reduction in the density of BLIMP1+ tumor cells, effector CD8+ T cells, and regulatory T cells, indicative of an altered immune microenvironment. Finally, bone heterogeneity decreased following treatment of patients with multiple myeloma. In summary, deep learning-based spatial mapping of bone marrow trephine biopsies can provide insights into the cellular topography of the myeloma marrow microenvironment and complement aspirate-based techniques.

SIGNIFICANCE

Spatial analysis of bone marrow trephine biopsies using histology, deep learning, and tailored algorithms reveals the bone marrow architectural heterogeneity and evolution during myeloma progression and treatment.

Tailored risk assessment and forecasting in intermittent claudication

BACKGROUND

Guidelines recommend cardiovascular risk reduction and supervised exercise therapy as the first line of treatment in intermittent claudication, but implementation challenges and poor patient compliance lead to significant variation in management and therefore outcomes. The development of a precise risk stratification tool is proposed through a machine-learning algorithm that aims to provide personalized outcome predictions for different management strategies.

METHODS

Feature selection was performed using the least absolute shrinkage and selection operator method. The model was developed using a bootstrapped sample based on patients with intermittent claudication from a vascular centre to predict chronic limb-threatening ischaemia, two or more revascularization procedures, major adverse cardiovascular events, and major adverse limb events. Algorithm performance was evaluated using the area under the receiver operating characteristic curve. Calibration curves were generated to assess the consistency between predicted and actual outcomes. Decision curve analysis was employed to evaluate the clinical utility. Validation was performed using a similar dataset.

RESULTS

The bootstrapped sample of 10 000 patients was based on 255 patients. The model was validated using a similar sample of 254 patients. The area under the receiver operating characteristic curves for risk of progression to chronic limb-threatening ischaemia at 2 years (0.892), risk of progression to chronic limb-threatening ischaemia at 5 years (0.866), likelihood of major adverse cardiovascular events within 5 years (0.836), likelihood of major adverse limb events within 5 years (0.891), and likelihood of two or more revascularization procedures within 5 years (0.896) demonstrated excellent discrimination. Calibration curves demonstrated good consistency between predicted and actual outcomes and decision curve analysis confirmed clinical utility. Logistic regression yielded slightly lower area under the receiver operating characteristic curves for these outcomes compared with the least absolute shrinkage and selection operator algorithm (0.728, 0.717, 0.746, 0.756, and 0.733 respectively). External calibration curve and decision curve analysis confirmed the reliability and clinical utility of the model, surpassing traditional logistic regression.

CONCLUSION

The machine-learning algorithm successfully predicts outcomes for patients with intermittent claudication across various initial treatment strategies, offering potential for improved risk stratification and patient outcomes.

Assessing the Utility of artificial intelligence in endometriosis: Promises and pitfalls

Endometriosis, a chronic condition characterized by the growth of endometrial-like tissue outside of the uterus, poses substantial challenges in terms of diagnosis and treatment. Artificial intelligence (AI) has emerged as a promising tool in the field of medicine, offering opportunities to address the complexities of endometriosis. This review explores the current landscape of endometriosis diagnosis and treatment, highlighting the potential of AI to alleviate some of the associated burdens and underscoring common pitfalls and challenges when employing AI algorithms in this context. Women's health research in endometriosis has suffered from underfunding, leading to limitations in diagnosis, classification, and treatment approaches. The heterogeneity of symptoms in patients with endometriosis has further complicated efforts to address this condition. New, powerful methods of analysis have the potential to uncover previously unidentified patterns in data relating to endometriosis. AI, a collection of algorithms replicating human decision-making in data analysis, has been increasingly adopted in medical research, including endometriosis studies. While AI offers the ability to identify novel patterns in data and analyze large datasets, its effectiveness hinges on data quality and quantity and the expertise of those implementing the algorithms. Current applications of AI in endometriosis range from diagnostic tools for ultrasound imaging to predicting treatment success. These applications show promise in reducing diagnostic delays, healthcare costs, and providing patients with more treatment options, improving their quality of life. AI holds significant potential in advancing the diagnosis and treatment of endometriosis, but it must be applied carefully and transparently to avoid pitfalls and ensure reproducibility. This review calls for increased scrutiny and accountability in AI research. Addressing these challenges can lead to more effective AI-driven solutions for endometriosis and other complex medical conditions.

Effect size and inferential statistical techniques coupled with machine learning for assessing the association between prolactin concentration and metabolic homeostasis

BACKGROUND AND OBJECTIVE

Recent guidelines classify low prolactin levels as low as <7 ng/mL and high levels as >25 ng/mL, while the "Homeostatically Functionally Increased Transient Prolactinemia" (HomeoFIT-PRL) range (25-100 ng/mL) suggests that a temporary increase in prolactin could be metabolically beneficial if no related health issues are present. The aim of this study was to investigate the association between mean prolactin concentrations and disturbances in glycidic and lipidic metabolism and to identify the gray zone associated with prolactin inflection points that correlate with these metabolic changes.

METHODS

This cross-sectional study involved 65,795 adults who underwent HOMA-IR, glucose, insulin, total cholesterol, HDL-c, LDL-c, and triglyceride tests. Data was categorized into 106 partitions based on prolactin results. Employing an approach referred to in this study as "Hierarchical Multicriteria Analysis of Differences Between Groups - Statistical and Effect Size Approach" (HiMADiG-SESA) comparing the mean concentrations of metabolic tests across prolactin ranges. A machine learning model was utilized to determine inflection points and their corresponding confidence intervals (CIs). These CIs helped establish gray zones in mean prolactin results related to metabolic changes.

RESULTS

Statistically and clinically, metabolic test means differed for prolactin <7 ng/mL, except insulin. In the HomeoFIT-PRL range, means were lower except for HDL-c. The gray zones of the mean prolactin results associated with changes in glycidic and lipidic metabolism were 9.58-12.87 ng/mL and 13.81-18.73 ng/mL, respectively.

CONCLUSION

A strong correlation was identified between mean prolactin concentrations and the results of metabolism tests below the gray zones associated with inflection points, indicating the potential role of prolactin in the appearance of metabolic disorders. Mean prolactin results can provide deeper insight into metabolic balance.

Biomarkers prediction and immune landscape in ulcerative colitis: Findings based on bioinformatics and machine learning

BACKGROUND

Ulcerative colitis (UC) presents diagnostic and therapeutic difficulties. The primary objective of this study is to identify efficacious biomarkers for diagnosis and treatment, as well as acquire a deeper understanding of the immuneological characteristics associated with the disease.

METHODS

Datasets relating to UC were obtained from GEO database. Among these, three datasets were merged to create a metadata for bioinformatics analysis and machine learning. Additionally, one dataset specifically utilized for external validation. Least absolute shrinkage and selection operator (LASSO) and random forest (RF) were employed to screen signature genes. The artificial neural network (ANN) model and receiver operating characteristic (ROC) curve were used to assess the diagnostic performance of signature genes. The single sample gene set enrichment analysis (ssGSEA) was applied to reveal the immune landscape. Finally, the relationship between the signature genes, immune infiltration, and clinical characteristics was investigated through correlation analysis.

RESULT

By intersecting the result of LASSO, RF and WGCNA, 8 signature genes were identified, including S100A8, IL-1B, CXCL1, TCN1, MMP10, GREM1, DUOX2 and SLC6A14. The biological progress of this gene mostly encompasses acute inflammatory response, aggregation and chemotaxis of leukocyte, and response to lipopolysaccharide by mediating IL-17 signaling pathway, NF-kappa B signaling pathway, TNF signaling pathway, NOD-like receptor signaling pathway. Immune infiltration analysis shows 25 immune cells are significantly elevated in UC samples. Moreover, these signature genes exhibit a strong correlation with various immune cells and a mild to moderate correlation with the Mayo score.

CONCLUSION

S100A8, IL-1B, CXCL1, TCN1, MMP10, GREM1, DUOX2 and SLC6A14 have been identified as credible potential biomarkers for the diagnosis and therapy of UC. The immune response mediated by these signature biomarkers plays a crucial role in the occurrence and advancement of UC by means of the reciprocal interaction between the signature biomarkers and immune-infiltrated cells.

Diagnostic Test Accuracy of artificial intelligence-assisted detection of acute coronary syndrome: A systematic review and meta-analysis

BACKGROUND

Artificial intelligence (AI) has potential uses in healthcare including the detection of health conditions and prediction of health outcomes. Past systematic reviews had reviewed the accuracy of artificial neural networks (ANN) on Electrocardiogram (ECG) readings but that of other AI models on other Acute Coronary Syndrome (ACS) detection tools remains unclear.

METHODS

Nine electronic databases were searched from 2012 to 31 August 2022 including grey literature search and hand searching of references of included articles. Risk of bias was assessed by two independent reviewers using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). Test characteristics namely true positives, false positives, true negatives, and false negatives were extracted from all included articles into a 2x2 table. Study-specific estimates of sensitivity and specificity were pooled using hierarchical summary receiver operating characteristic (HSROC) model and displayed using a forest plot and HSROC curve.

RESULTS

66 studies were included in the review. A total of 518,931 patients were included whose mean ages varied from 32.62 to 70 years old. In 66 studies, the sensitivity and specificity of AI-based detection for ACS screening ranged from 64 % to 100 % and 65 %-100 %, respectively. The overall quality of evidence was low due to the inclusion of case-control studies.

CONCLUSION

Results of the study inform the potential of using AI-assisted ACS detection for accurate diagnosis and prompt treatment for ACS. Adherence to the Standards for Reporting of Diagnostic Accuracy (STARD) guideline and having more cohort studies for future Diagnostic Test Accuracy (DTA) studies are necessary to improve the quality of evidence of AI-based detection of ACS.

A prognostic model and pre-discharge predictors of post-COVID-19 syndrome after hospitalization for SARS-CoV-2 infection

Background

Post-COVID-19 syndrome (PCS) has been increasingly recognized as an emerging problem: 50% of patients report ongoing symptoms 1 year after acute infection, with most typical manifestations (fatigue, dyspnea, psychiatric and neurological symptoms) having potentially debilitating effect. Early identification of high-risk candidates for PCS development would facilitate the optimal use of resources directed to rehabilitation of COVID-19 convalescents.

Objective

To study the in-hospital clinical characteristics of COVID-19 survivors presenting with self-reported PCS at 3 months and to identify the early predictors of its development.

Methods

221 hospitalized COVID-19 patients underwent symptoms assessment, 6-min walk test, and echocardiography pre-discharge and at 1 month; presence of PCS was assessed 3 months after discharge. Unsupervised machine learning was used to build a SANN-based binary classification model of PCS development.

Results

PCS at 3 months has been detected in 75% patients. Higher symptoms level in the PCS group was not associated with worse physical functional recovery or significant echocardiographic changes. Despite identification of a set of pre-discharge predictors, inclusion of parameters obtained at 1 month proved necessary to obtain a high accuracy model of PCS development, with inputs list including age, sex, in-hospital levels of CRP, eGFR and need for oxygen supplementation, and level of post-exertional symptoms at 1 month after discharge (fatigue and dyspnea in 6MWT and MRC Dyspnea score).

Conclusion

Hospitalized COVID-19 survivors at 3 months were characterized by 75% prevalence of PCS, the development of which could be predicted with an 89% accuracy using the derived neural network-based classification model.

From immunology to artificial intelligence: revolutionizing latent tuberculosis infection diagnosis with machine learning

Latent tuberculosis infection (LTBI) has become a major source of active tuberculosis (ATB). Although the tuberculin skin test and interferon-gamma release assay can be used to diagnose LTBI, these methods can only differentiate infected individuals from healthy ones but cannot discriminate between LTBI and ATB. Thus, the diagnosis of LTBI faces many challenges, such as the lack of effective biomarkers from Mycobacterium tuberculosis (MTB) for distinguishing LTBI, the low diagnostic efficacy of biomarkers derived from the human host, and the absence of a gold standard to differentiate between LTBI and ATB. Sputum culture, as the gold standard for diagnosing tuberculosis, is time-consuming and cannot distinguish between ATB and LTBI. In this article, we review the pathogenesis of MTB and the immune mechanisms of the host in LTBI, including the innate and adaptive immune responses, multiple immune evasion mechanisms of MTB, and epigenetic regulation. Based on this knowledge, we summarize the current status and challenges in diagnosing LTBI and present the application of machine learning (ML) in LTBI diagnosis, as well as the advantages and limitations of ML in this context. Finally, we discuss the future development directions of ML applied to LTBI diagnosis.

The prediction of sagittal chin point relapse following two-jaw surgery using machine learning

The study aimed to identify critical factors associated with the surgical stability of pogonion (Pog) by applying machine learning (ML) to predict relapse following two-jaw orthognathic surgery (2 J-OGJ). The sample set comprised 227 patients (110 males and 117 females, 207 training and 20 test sets). Using lateral cephalograms taken at the initial evaluation (T0), pretreatment (T1), after (T2) 2 J-OGS, and post treatment (T3), 55 linear and angular skeletal and dental surgical movements (T2-T1) were measured. Six ML modes were utilized, including classification and regression trees (CART), conditional inference tree (CTREE), and random forest (RF). The training samples were classified into three groups; highly significant (HS) (≥ 4), significant (S) (≥ 2 and < 4), and insignificant (N), depending on Pog relapse. RF indicated that the most important variable that affected relapse rank prediction was ramus inclination (RI), CTREE and CART revealed that a clockwise rotation of more than 3.7 and 1.8 degrees of RI was a risk factor for HS and S groups, respectively. RF, CTREE, and CART were practical tools for predicting surgical stability. More than 1.8 degrees of CW rotation of the ramus during surgery would lead to significant Pog relapse.

CT radiomics can predict disease progression within 6 months after chimeric antigen receptor-modified T-cell therapy in relapsed/refractory B-cell non-Hodgkin's lymphoma patients

AIM

To predict progression within 6 months after chimeric antigen receptor-modified (CAR) T-cell therapy for relapsed/refractory (R/R) B-cell non-Hodgkin's lymphoma (B-NHL) patients by radiomic indexes derived from contrast-enhanced computed tomography (CECT) examinations.

MATERIALS AND METHODS

Seventy R/R B-NHL patients who underwent CECT before treatment with CAR T-cells were examined retrospectively. In total, 297 volumes of interest for lesions were segmented from CECT images. Patients without and with disease progression were assigned to groups 1 and 2, respectively. Radiomic and combined predictive models were constructed by three machine-learning algorithms using features from the training set, respectively. Furthermore, predictive models were constructed based on multi-lesion-based and largest-lesion-based radiomic features, respectively.

RESULTS

In the test set, no marked differences were observed between the areas under the curves (AUCs) of the combined and radiomic models for all three machine-learning algorithms (all p>0.05). Differences in machine-learning algorithms did not significantly affect the predictive performances of the models. Radiomics and combined models constructed with multi-lesion-based radiomic features showed better predictive performances than those applying largest-lesion-based radiomic features (all p<0.05 for comparisons between combined models).

CONCLUSION

CECT-based radiomic features may be applied to predict disease progression in R/R B-NHL patients within 6 months after CAR T-cell treatment, and radiomic features from multiple lesions may have better predictive efficacy. Different machine-learning algorithms may not show significant differences in prediction performance.

A radiomics-based logistic regression model for the assessment of emphysema severity

Introduction

The aim of this study is to develop a model that differentiates between the radiological patterns of severe and mild emphysema using radiomics parameters, as well as to examine the parameters included in the model.

Materials and Methods

Over the last 12 months, a total of 354 patients were screened based on the presence of terms such as “Fleischner”, “CLE”, and “centriacinar” in their thoracic CT reports, culminating in a study population of 82 patients. The study population was divided into Group 1 (Fleischner mild and moderate; n= 45) and Group 2 (Fleischner confluent and advanced destructive; n= 37). Volumetric segmentation was performed, focusing on the upper lobe segments of both lungs. From these segmented volumes, radiomics parameters including shape, size, first-order, and second-order features were calculated. The best model parameters were selected based on the Bayesian Information Criterion and further optimized through grid search. The final model was tested using 1000 iterations of bootstrap resampling.

Results

In the training set, performance metrics were calculated with a sensitivity of 0.862, specificity of 0.870, accuracy of 0.863, and AUC of 0.910. Correspondingly, in the test set, these values were sensitivity= 0.848; specificity= 0.865; accuracy= 0.857; and AUC= 0.907.

Conclusion

The logistic regression model, composed of radiomics parameters and trained on a limited number of cases, effectively differentiated between mild and severe radiological patterns of emphysema using computed tomography images.

Machine learning-based risk prediction model for canine myxomatous mitral valve disease using electronic health record data

Introduction

Myxomatous mitral valve disease (MMVD) is the most common cause of heart failure in dogs, and assessing the risk of heart failure in dogs with MMVD is often challenging. Machine learning applied to electronic health records (EHRs) is an effective tool for predicting prognosis in the medical field. This study aimed to develop machine learning-based heart failure risk prediction models for dogs with MMVD using a dataset of EHRs.

Methods

A total of 143 dogs with MMVD between May 2018 and May 2022. Complete medical records were reviewed for all patients. Demographic data, radiographic measurements, echocardiographic values, and laboratory results were obtained from the clinical database. Four machine-learning algorithms (random forest, K-nearest neighbors, naïve Bayes, support vector machine) were used to develop risk prediction models. Model performance was represented by plotting the receiver operating characteristic (ROC) curve and calculating the area under the curve (AUC). The best-performing model was chosen for the feature-ranking process.

Results

The random forest model showed superior performance to the other models (AUC = 0.88), while the performance of the K-nearest neighbors model showed the lowest performance (AUC = 0.69). The top three models showed excellent performance (AUC ≥ 0.8). According to the random forest algorithm's feature ranking, echocardiographic and radiographic variables had the highest predictive values for heart failure, followed by packed cell volume (PCV) and respiratory rates. Among the electrolyte variables, chloride had the highest predictive value for heart failure.

Discussion

These machine-learning models will enable clinicians to support decision-making in estimating the prognosis of patients with MMVD.

A Machine Learning-Based Analytic Pipeline Applied to Clinical and Serum IgG Immunoproteome Data To Predict Chlamydia trachomatis Genital Tract Ascension and Incident Infection in Women

We developed a reusable and open-source machine learning (ML) pipeline that can provide an analytical framework for rigorous biomarker discovery. We implemented the ML pipeline to determine the predictive potential of clinical and immunoproteome antibody data for outcomes associated with Chlamydia trachomatis (Ct) infection collected from 222 cis-gender females with high Ct exposure. We compared the predictive performance of 4 ML algorithms (naive Bayes, random forest, extreme gradient boosting with linear booster [xgbLinear], and k-nearest neighbors [KNN]), screened from 215 ML methods, in combination with two different feature selection strategies, Boruta and recursive feature elimination. Recursive feature elimination performed better than Boruta in this study. In prediction of Ct ascending infection, naive Bayes yielded a slightly higher median value of are under the receiver operating characteristic curve (AUROC) 0.57 (95% confidence interval [CI], 0.54 to 0.59) than other methods and provided biological interpretability. For prediction of incident infection among women uninfected at enrollment, KNN performed slightly better than other algorithms, with a median AUROC of 0.61 (95% CI, 0.49 to 0.70). In contrast, xgbLinear and random forest had higher predictive performances, with median AUROC of 0.63 (95% CI, 0.58 to 0.67) and 0.62 (95% CI, 0.58 to 0.64), respectively, for women infected at enrollment. Our findings suggest that clinical factors and serum anti-Ct protein IgGs are inadequate biomarkers for ascension or incident Ct infection. Nevertheless, our analysis highlights the utility of a pipeline that searches for biomarkers and evaluates prediction performance and interpretability. IMPORTANCE Biomarker discovery to aid early diagnosis and treatment using machine learning (ML) approaches is a rapidly developing area in host-microbe studies. However, lack of reproducibility and interpretability of ML-driven biomarker analysis hinders selection of robust biomarkers that can be applied in clinical practice. We thus developed a rigorous ML analytical framework and provide recommendations for enhancing reproducibility of biomarkers. We emphasize the importance of robustness in selection of ML methods, evaluation of performance, and interpretability of biomarkers. Our ML pipeline is reusable and open-source and can be used not only to identify host-pathogen interaction biomarkers but also in microbiome studies and ecological and environmental microbiology research.

Spontaneous physical functional recovery after hospitalization for COVID-19: insights from a 1 month follow-up and a model to predict poor trajectory

Background

Long COVID syndrome has emerged as a new global healthcare challenge, with impaired physical performance being a prominent debilitating factor. Cardiopulmonary rehabilitation is a mainstay of management of symptomatic post-COVID patients, and optimization of candidate selection might allow for more effective use of available resources.

Methods

In order to study the natural dynamics and to identify predictors of physical functional recovery following hospitalization for COVID-19, 6 min walk test was performed pre-discharge in 176 patients (40% hypertensive, 53% female, mean age 53.2 ± 13.5 years) with re-evaluation at 1 month.

Results

Six min walk distance and the reached percent of predicted distance (6MWD%) were suboptimal at both visits-396 ± 71 m (68.7 ± 12.4%) pre-discharge and 466 ± 65 m (81.8 ± 13.6%) at 1 month. Associated changes included significant oxygen desaturation (2.9 ± 2.5 and 2.3 ± 2.2%, respectively) and insufficient increment of heart rate during the test (24.9 ± 17.5 and 28.2 ± 12.0 bpm) that resulted in low reached percent of individual maximum heart rate (61.1 ± 8.1 and 64.3 ± 8.2%). Automatic clusterization of the study cohort by the 6MWD% changes has allowed to identify the subgroup of patients with poor "low base-low increment" trajectory of spontaneous post-discharge recovery that were characterized by younger age (38.2 ± 11.0 vs. 54.9 ± 12.1, p < 0.001) but more extensive pulmonary involvement by CT (43.7 ± 8.8 vs. 29.6 ± 19.4%, p = 0.029) and higher peak ESR values (36.5 ± 9.7 vs. 25.6 ± 12.8, p < 0.001). Predictors of poor recovery in multivariate logistic regression analysis included age, peak ESR, eGFR, percentage of pulmonary involvement by CT, need for in-hospital oxygen supplementation, SpO2 and mMRC dyspnea score pre-discharge, and history of hypertension.

Conclusion

COVID-19 survivors were characterized by decreased physical performance pre-discharge as assessed by the 6 min walk test and did not completely restore their functional status after 1 month of spontaneous recovery, with signs of altered blood oxygenation and dysautonomia contributing to the observed changes. Patients with poor "low base-low increment" trajectory of post-discharge recovery were characterized by younger age but more extensive pulmonary involvement and higher peak ESR values. Poor post-discharge recovery in the study cohort was predictable by the means of machine learning-based classification model that used age, history of hypertension, need for oxygen supplementation, and ESR as inputs.

Canine Mammary Tumor Histopathological Image Classification via Computer-Aided Pathology: An Available Dataset for Imaging Analysis

Histopathology, the gold-standard technique in classifying canine mammary tumors (CMTs), is a time-consuming process, affected by high inter-observer variability. Digital (DP) and Computer-aided pathology (CAD) are emergent fields that will improve overall classification accuracy. In this study, the ability of the CAD systems to distinguish benign from malignant CMTs has been explored on a dataset-namely CMTD-of 1056 hematoxylin and eosin JPEG images from 20 benign and 24 malignant CMTs, with three different CAD systems based on the combination of a convolutional neural network (VGG16, Inception v3, EfficientNet), which acts as a feature extractor, and a classifier (support vector machines (SVM) or stochastic gradient boosting (SGB)), placed on top of the neural net. Based on a human breast cancer dataset (i.e., BreakHis) (accuracy from 0.86 to 0.91), our models were applied to the CMT dataset, showing accuracy from 0.63 to 0.85 across all architectures. The EfficientNet framework coupled with SVM resulted in the best performances with an accuracy from 0.82 to 0.85. The encouraging results obtained by the use of DP and CAD systems in CMTs provide an interesting perspective on the integration of artificial intelligence and machine learning technologies in cancer-related research.