Understanding Bias in Artificial Intelligence: A Practice Perspective

In the fall of 2021, several experts in this space delivered a Webinar hosted by the American Society of Neuroradiology (ASNR) Diversity and Inclusion Committee, focused on expanding the understanding of bias in artificial intelligence, with a health equity lens, and provided key concepts for neuroradiologists to approach the evaluation of these tools. In this perspective, we distill key parts of this discussion, including understanding why this topic is important to neuroradiologists and lending insight on how neuroradiologists can develop a framework to assess health equity-related bias in artificial intelligence tools. In addition, we provide examples of clinical workflow implementation of these tools so that we can begin to see how artificial intelligence tools will impact discourse on equitable radiologic care. As continuous learners, we must be engaged in new and rapidly evolving technologies that emerge in our field. The Diversity and Inclusion Committee of the ASNR has addressed this subject matter through its programming content revolving around health equity in neuroradiologic advances.

Application of novel PACS-based informatics platform to identify imaging based predictors of CDKN2A allelic status in glioblastomas

Gliomas with CDKN2A mutations are known to have worse prognosis but imaging features of these gliomas are unknown. Our goal is to identify CDKN2A specific qualitative imaging biomarkers in glioblastomas using a new informatics workflow that enables rapid analysis of qualitative imaging features with Visually AcceSAble Rembrandtr Images (VASARI) for large datasets in PACS. Sixty nine patients undergoing GBM resection with CDKN2A status determined by whole-exome sequencing were included. GBMs on magnetic resonance images were automatically 3D segmented using deep learning algorithms incorporated within PACS. VASARI features were assessed using FHIR forms integrated within PACS. GBMs without CDKN2A alterations were significantly larger (64 vs. 30%, p = 0.007) compared to tumors with homozygous deletion (HOMDEL) and heterozygous loss (HETLOSS). Lesions larger than 8 cm were four times more likely to have no CDKN2A alteration (OR: 4.3; 95% CI 1.5-12.1; p < 0.001). We developed a novel integrated PACS informatics platform for the assessment of GBM molecular subtypes and show that tumors with HOMDEL are more likely to have radiographic evidence of pial invasion and less likely to have deep white matter invasion or subependymal invasion. These imaging features may allow noninvasive identification of CDKN2A allele status.

Systematic Literature Review of Machine Learning Algorithms Using Pretherapy Radiologic Imaging for Glioma Molecular Subtype Prediction

BACKGROUND

The molecular profile of gliomas is a prognostic indicator for survival, driving clinical decision-making for treatment. Pathology-based molecular diagnosis is challenging because of the invasiveness of the procedure, exclusion from neoadjuvant therapy options, and the heterogeneous nature of the tumor.

PURPOSE

We performed a systematic review of algorithms that predict molecular subtypes of gliomas from MR Imaging.

DATA SOURCES

Data sources were Ovid Embase, Ovid MEDLINE, Cochrane Central Register of Controlled Trials, Web of Science.

STUDY SELECTION

Per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 12,318 abstracts were screened and 1323 underwent full-text review, with 85 articles meeting the inclusion criteria.

DATA ANALYSIS

We compared prediction results from different machine learning approaches for predicting molecular subtypes of gliomas. Bias analysis was conducted for each study, following the Prediction model Risk Of Bias Assessment Tool (PROBAST) guidelines.

DATA SYNTHESIS

Isocitrate dehydrogenase mutation status was reported with an area under the curve and accuracy of 0.88 and 85% in internal validation and 0.86 and 87% in limited external validation data sets, respectively. For the prediction of O6-methylguanine-DNA methyltransferase promoter methylation, the area under the curve and accuracy in internal validation data sets were 0.79 and 77%, and in limited external validation, 0.89 and 83%, respectively. PROBAST scoring demonstrated high bias in all articles.

LIMITATIONS

The low number of external validation and studies with incomplete data resulted in unequal data analysis. Comparing the best prediction pipelines of each study may introduce bias.

CONCLUSIONS

While the high area under the curve and accuracy for the prediction of molecular subtypes of gliomas are reported in internal and external validation data sets, limited use of external validation and the increased risk of bias in all articles may present obstacles for clinical translation of these techniques.

The Brain Tumor Segmentation (BraTS-METS) Challenge 2023: Brain Metastasis Segmentation on Pre-treatment MRI

Clinical monitoring of metastatic disease to the brain can be a laborious and timeconsuming process, especially in cases involving multiple metastases when the assessment is performed manually. The Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) guideline, which utilizes the unidimensional longest diameter, is commonly used in clinical and research settings to evaluate response to therapy in patients with brain metastases. However, accurate volumetric assessment of the lesion and surrounding peri-lesional edema holds significant importance in clinical decision-making and can greatly enhance outcome prediction. The unique challenge in performing segmentations of brain metastases lies in their common occurrence as small lesions. Detection and segmentation of lesions that are smaller than 10 mm in size has not demonstrated high accuracy in prior publications. The brain metastases challenge sets itself apart from previously conducted MICCAI challenges on glioma segmentation due to the significant variability in lesion size. Unlike gliomas, which tend to be larger on presentation scans, brain metastases exhibit a wide range of sizes and tend to include small lesions. We hope that the BraTS-METS dataset and challenge will advance the field of automated brain metastasis detection and segmentation.

Application of artificial intelligence centric workflows for evaluation of neuroradiology emergencies

The goal of this study was to perform a pilot study to assess user-interface of radiologists with an artificial-intelligence (AI) centric workflow for detection of intracranial hemorrhage (ICH) and cervical spine fractures (CSFX). Over 12-month period, interaction and usage of AI software implemented in our institution, Aidoc, on head and cervical spine CT scans were obtained. Several interaction variables were defined, assessing different types of interaction between readers of different training level and AI software. The median usage of AI-centric workflow for detection of ICH and CSFX were 28.8% and 21.8%, respectively, demonstrating a significant additional engagement beyond Native workflow (worklist and PACS). Further studies are warranted to expand interaction assessments to further understand the value unlocked by the AI-centric workflows.

Approaches to Greening Radiology

The health care sector is a resource-intensive industry, consuming significant amounts of water and energy, and producing a multitude of waste. Health care providers are increasingly implementing strategies to reduce energy use and waste. Little is currently known about existing sustainability strategies and how they may be supported by radiology practices. Here, we review concepts and ideas that minimize energy use and waste, and that can be supported or implemented by radiologists.

Evolution and implementation of radiographic response criteria in neuro-oncology

Radiographic response assessment in neuro-oncology is critical in clinical practice and trials. Conventional criteria, such as the MacDonald and response assessment in neuro-oncology (RANO) criteria, rely on bidimensional (2D) measurements of a single tumor cross-section. Although RANO criteria are established for response assessment in clinical trials, there is a critical need to address the complexity of brain tumor treatment response with multiple new approaches being proposed. These include volumetric analysis of tumor compartments, structured MRI reporting systems like the Brain Tumor Reporting and Data System, and standardized approaches to advanced imaging techniques to distinguish tumor response from treatment effects. In this review, we discuss the strengths and limitations of different neuro-oncology response criteria and summarize current research findings on the role of novel response methods in neuro-oncology clinical trials and practice.

Acute Bacterial Skin and Skin-Structure Infections, efficacy of Dalbavancin: a systematic review and meta-analysis

OBJECTIVES

To know the efficacy of different doses of dalbavancin in acute bacterial skin and skin-structure infections (ABSSSIs) and versus other antibiotics.

METHODS

We performed a systematic review of dalbavancin efficacy for ABSSSIs. We selected 10 clinical trials from MEDLINE and Cochrane databases for qualitative review. Of these, five trials compared one or two doses of dalbavancin versus other antibiotics such as vancomycin or linezolid.

RESULTS

Treatment outcomes with other antibiotics were not significantly different versus two doses of dalbavancin (OR 1.13; 95% CI 0.75-1.71; p = 0.55) or single dose dalbavancin (OR 0.98; 95% CI 0.19-5.17; p = 0.98). One dose versus two doses of dalbavancin did not show significant differences in any of the treatment groups. In contrast, the global microbiological assessment results indicated a favorable outcome for two doses of dalbavancin compared to the single dose of dalbavancin (OR 2.96; 95% CI 1.19-7.39; p = 0.02) in both methicillin-resistant and methicillin-susceptible Staphylococcus aureus.

CONCLUSION

Either single dose or two dose dalbavancin treatment is as clinically effective as other antibiotics such as vancomycin and linezolid for the treatment of ABSSSIs.Abbreviations ABSSI: acute bacterial skin and skin-structure infection; AUC: area under the concentration-time curve; CE: clinical evaluable; CI: confidence interval; EOT: end of treatment; ITT: intention-to-treat; LOS: length of stay; MIC: minimum inhibitory concentration; MIC₉₀: minimum concentration to inhibit growth of 90% of isolates; MR: methicillin resistant; MRSA: methicillin-resistant Staphylococcus aureus; MS: methicillin susceptible; MSSA: methicillin-susceptible Staphylococcus aureus; OPAT: Outpatient Parenteral Antimicrobial Therapy; OR: odds ratio; PI: penicillin intermediate; PR: penicillin resistant; PS penicillin susceptible; SIRS: systemic inflammatory response syndrome; SSTI: skin and soft tissue infection; TOC: test of cure; VR: vancomycin resistant; VS: vancomycin susceptible.

Global trends of dalbavancin: A bibliometric analysis

WHAT IS KNOWN AND OBJECTIVE?: Dalbavancin is used against gram-positive pathogens such as Staphylococcus aureus in acute bacterial skin and skin-structure infections. METHODS: Our main goal was to identify the key articles sustaining the current knowledge of this drug's therapeutic possibilities through a bibliometric analysis of the available literature. RESULTS AND DISCUSSION: On 15 March 2021, we searched the Web of Science electronically for documents that contain within its title the term "dalbavancin." We found a total of 675 documents that average 20.23 citations/publication with a density of 682.60 citations per/year, yielding an h-index of 58. After ranking them by the number of times cited, we extracted the top 100 most-cited records (T100). Number of citations/publication ranged from 13 to 231, publication years were 2002-2019, with the top-cited article published in 2014. All T100 publications were written in English. JMI Laboratories was the institution with the most articles in the T100 (22 documents), and the United States was the top country (75 documents). Five authors participated in at least five of the T100, led by Jones RN with 20 articles. Positions #1, #2, #5, and #9 were clinical trials for acute bacterial skin and skin structure infections (ABSSSI), the on-label indication for dalbavancin. Only one article in the top 10 (T10) was an off-label indication that was published in 2005 with 186 citations, and occupied the third position among the T100. Using the VOSviewer© programme, we observed that the most used keywords were: dalbavancin, lipoglycopeptide, gram-positive, osteomyelitis, vancomycin, and MRSA. WHAT IS NEW AND CONCLUSIONS?: Our study identifies the most significant research on dalbavancin, including the highest impact publications, and highlights the recent trend of dalbavancin in new therapies. The T10 articles include the most important dalbavancin clinical trials, along with other studies and reviews that support the growing role of this antibiotic in clinical use. Emphasis has been on the favourable pharmacokinetic profile that allows administration once-weekly, with minimal risk of severe adverse events.

Identifying clinically applicable machine learning algorithms for glioma segmentation: recent advances and discoveries

Background

While there are innumerable machine learning (ML) research algorithms used for segmentation of gliomas, there is yet to be a US FDA cleared product. The aim of this study is to explore the systemic limitations of research algorithms that have prevented translation from concept to product by a review of the current research literature.

Methods

We performed a systematic literature review on 4 databases. Of 11 727 articles, 58 articles met the inclusion criteria and were used for data extraction and screening using TRIPOD.

Results

We found that while many articles were published on ML-based glioma segmentation and report high accuracy results, there were substantial limitations in the methods and results portions of the papers that result in difficulty reproducing the methods and translation into clinical practice.

Conclusions

In addition, we identified that more than a third of the articles used the same publicly available BRaTS and TCIA datasets and are responsible for the majority of patient data on which ML algorithms were trained, which leads to limited generalizability and potential for overfitting and bias.

Machine Learning Tools for Image-Based Glioma Grading and the Quality of Their Reporting: Challenges and Opportunities

Technological innovation has enabled the development of machine learning (ML) tools that aim to improve the practice of radiologists. In the last decade, ML applications to neuro-oncology have expanded significantly, with the pre-operative prediction of glioma grade using medical imaging as a specific area of interest. We introduce the subject of ML models for glioma grade prediction by remarking upon the models reported in the literature as well as by describing their characteristic developmental workflow and widely used classifier algorithms. The challenges facing these models-including data sources, external validation, and glioma grade classification methods -are highlighted. We also discuss the quality of how these models are reported, explore the present and future of reporting guidelines and risk of bias tools, and provide suggestions for the reporting of prospective works. Finally, this review offers insights into next steps that the field of ML glioma grade prediction can take to facilitate clinical implementation.

Real-time PACS-integrated longitudinal brain metastasis tracking tool provides comprehensive assessment of treatment response to radiosurgery

Background

Treatment of brain metastases can be tailored to individual lesions with treatments such as stereotactic radiosurgery. Accurate surveillance of lesions is a prerequisite but challenging in patients with multiple lesions and prior imaging studies, in a process that is laborious and time consuming. We aimed to longitudinally track several lesions using a PACS-integrated lesion tracking tool (LTT) to evaluate the efficiency of a PACS-integrated lesion tracking workflow, and characterize the prevalence of heterogenous response (HeR) to treatment after Gamma Knife (GK).

Methods

We selected a group of brain metastases patients treated with GK at our institution. We used a PACS-integrated LTT to track the treatment response of each lesion after first GK intervention to maximally seven diagnostic follow-up scans. We evaluated the efficiency of this tool by comparing the number of clicks necessary to complete this task with and without the tool and examined the prevalence of HeR in treatment.

Results

A cohort of eighty patients was selected and 494 lesions were measured and tracked longitudinally for a mean follow-up time of 374 days after first GK. Use of LTT significantly decreased number of necessary clicks. 81.7% of patients had HeR to treatment at the end of follow-up. The prevalence increased with increasing number of lesions.

Conclusions

Lesions in a single patient often differ in their response to treatment, highlighting the importance of individual lesion size assessments for further treatment planning. PACS-integrated lesion tracking enables efficient lesion surveillance workflow and specific and objective result reports to treating clinicians.

Sinonasal Glomangiopericytoma: Review of Imaging Appearance and Clinical Management Update for a Rare Sinonasal Neoplasm

Introduction: Glomangiopericytoma (GPC) is a rare tumor in the nasal cavity or paranasal sinuses with low malignant potential. Initially deemed a hemangiopericytoma, in 2005 it was classified as a distinct entity by the World Health Organization (WHO). Case Presentation: A male patient in his early 60s presented with new-onset right arm and leg weakness/numbness, who was incidentally found to have a left ethmoid sinus mass with extension in the olfactory fossa. On CT and MRI, the mass enhanced with well-defined borders and eroded the bone, but without dural enhancement. The mass was surgically excised, and pathology confirmed the diagnosis of glomangiopericytoma by microscopic appearance and staining. Discussion: Glomangiopericytoma has less than 0.5% incidence of all neoplasms of the sinonasal cavity, making it rare. Most diagnosed patients are in their 6th or 7th decade of age, with a slight female predominance. Treatment is complete surgical excision, with excellent prognosis, although there is up to 17% local recurrence. Despite the non-specific appearance on CT and MRI, imaging can help provide differential diagnosis, tumor extent, size, and reassuring non-aggressive characteristics of the tumor prior to surgery. GPC tumors are relatively resistant to radiation and chemotherapy. Conclusion: It is important to recognize glomangiopericytoma in the differential of masses of the nasal cavities or paranasal sinuses, as they rarely warrant aggressive treatment beyond local excision. Each reported case of glomangiopericytoma helps to build guidance for imaging and treatment since GPC is rare and not well-represented in the medical literature.

A Case of Spinal Infectious Osteomyelitis Versus Gout: Advanced Imaging with Dual Energy CT

A 67-year-old male presented to the hospital for lower back pain and left lower extremity radiculopathy. Although the patient was afebrile and white blood cell count was normal, MRI was concerning for discitis/osteomyelitis at L4-L5. Subsequently, the patient developed a right knee joint effusion and underwent an arthrocentesis that was notable for the presence of urate crystals. A systemic urate crystal arthropathy was proposed as a potential etiology for the patient's back pain and radiculopathy. Dual energy CT of the lumbar spine was performed, a technique which determines material composition by comparing the photon attenuation of the substance from two different x-ray energy levels. Results revealed the presence of monosodium urate crystals in the intervertebral discs. This technique is proposed as a noninvasive way to evaluate for gout in atypical locations or those difficult to sample and may replace an invasive intervertebral disc/endplate aspiration and/or biopsy. Dual energy CT should be considered in patients with elevated serum uric acid and concern for spinal involvement of gout.

NIMG-46. SYSTEMATIC LITERATURE REVIEW OF ARTIFICIAL INTELLIGENCE ALGORITHMS USING PRE-THERAPY MR IMAGING FOR GLIOMA MOLECULAR SUBTYPE CLASSIFICATION

PURPOSE

Identifying molecular subtypes in gliomas has prognostic and therapeutic value, traditionally after invasive neurosurgical tumor resection or biopsy. Recent advances using artificial intelligence (AI) show promise in using pre-therapy imaging for predicting molecular subtype. We performed a systematic review of recent literature on AI methods used to predict molecular subtypes of gliomas.

METHODS

Literature review conforming to PRSIMA guidelines was performed for publications prior to February 2021 using 4 databases: Ovid Embase, Ovid MEDLINE, Cochrane trials (CENTRAL), and Web of Science core-collection. Keywords included: artificial intelligence, machine learning, deep learning, radiomics, magnetic resonance imaging, glioma, and glioblastoma. Non-machine learning and non-human studies were excluded. Screening was performed using Covidence software. Bias analysis was done using TRIPOD guidelines.

RESULTS

11,727 abstracts were retrieved. After applying initial screening exclusion criteria, 1,135 full text reviews were performed, with 82 papers remaining for data extraction. 57% used retrospective single center hospital data, 31.6% used TCIA and BRATS, and 11.4% analyzed multicenter hospital data. An average of 146 patients (range 34-462 patients) were included. Algorithms predicting IDH status comprised 51.8% of studies, MGMT 18.1%, and 1p19q 6.0%. Machine learning methods were used in 71.4%, deep learning in 27.4%, and 1.2% directly compared both methods. The most common algorithm for machine learning were support vector machine (43.3%), and for deep learning convolutional neural network (68.4%). Mean prediction accuracy was 76.6%.

CONCLUSION

Machine learning is the predominant method for image-based prediction of glioma molecular subtypes. Major limitations include limited datasets (60.2% with under 150 patients) and thus limited generalizability of findings. We recommend using larger annotated datasets for AI network training and testing in order to create more robust AI algorithms, which will provide better prediction accuracy to real world clinical datasets and provide tools that can be translated to clinical practice.

NIMG-71. IDENTIFYING CLINICALLY APPLICABLE MACHINE LEARNING ALGORITHMS FOR GLIOMA SEGMENTATION USING A SYSTEMATIC LITERATURE REVIEW

PURPOSE

Nowadays Machine learning (ML) algorithms are often used for segmentation of gliomas, but which algorithms provide the most accurate method for implementation into clinical practice has not fully been identified. We performed a systematic review of the literature to characterize the methods used for glioma segmentation and their accuracy.

METHODS

In accordance to PRISMA, a literature review was performed on four databases, Ovid Embase, Ovid MEDLINE, Cochrane trials (CENTRAL) and Web of science core-collection first in October 2020 and in February 2021. Keywords and controlled vocabulary included artificial intelligence, machine learning, deep learning, radiomics, magnetic resonance imaging, glioma, and glioblastoma. Publications were screened in Covidence and the bias analysis was done in agreement with TRIPOD.

RESULTS

Sixty-six articles were used for data extraction. BRATS and TCIA datasets were used in 36.6% of all studies, with average number of patients being 141 (range: 1 to 622). ML methods represented 45.3% of studies, with deep learning used in 54.7%; Dice score for the tumor core ranged from 0.72 to 0.95. The most common algorithm used in the machine learning papers was support vector machines (SVM) and for deep learning papers, it was Convolutional Neural Networks (CNN). Preliminary TRIPOD analysis yielded an average score from 12 (range: 7-16) with the majority of papers demonstrating deficiencies in description of the ML algorithm, funding role, data acquisition and measures of model performance.

CONCLUSION

In the last years, many articles were published on segmentation of gliomas using machine learning, thus establishing this method for tumor segmentation with high accuracy. However, the major limitations for clinically applicable use of ML in glioma segmentation include more than one-third of publications use the same datasets, thus limiting generalizability, increase the likelihood of overfitting, show and lack of ML network description and standardization in accuracy reporting.

NIMG-38. MEASURING ADHERENCE TO TRIPOD OF ARTIFICIAL INTELLIGENCE PAPERS IN THE GLIOMA SEGMENTATION

PURPOSE

Generalizability, reproducibility and objectivity are critical elements that need to be considered when translating machine learning models into clinical practice. While a large body of literature has been published on machine learning methods for segmentation of brain tumors, a systematic evaluation of paper quality and reproducibility has not been done. We investigated the use of “Transparent Reporting of studies on prediction models for Individual Prognosis Or Diagnosis” (TRIPOD) items, among papers published in this relatively new and growing field.

METHODS

According to PRISMA a literature review was performed on four databases, Ovid Embase, Ovid MEDLINE, Cochrane trials (CENTRAL) and Web of science core-collection first in October 2020 and a second time in February 2021. Keywords and controlled vocabulary included artificial intelligence, machine learning, deep learning, radiomics, magnetic resonance imaging, glioma, and glioblastoma. The publications were assessed in order to the TRIPOD items.

RESULTS

37 publications from our database search were screened in TRIPOD and yielded an average score of 12.08 with the maximum score being 16 and the minimum score 7. The best scoring item was interpretation (item 19) where all papers scored a point. The lowest scoring items were the title, the abstract, risk groups and the model performance (items number 1, 2, 11 and 16), where no paper scored a point. Less than 1% of the papers discussed the problem of missing data (item 9) and the funding of research (item 22).

CONCLUSION

TRIPOD analysis showed that a majority of the papers do not score high on critical elements that allow reproducibility, translation, and objectivity of research. An average score of 12.08 (40%) indicates that the publications usually achieve a relatively low score. The categories that were consistently poorly described include the ML network description, measuring model performance, title details and inclusion of information into the abstract.

Review of Artificial Intelligence Training Tools and Courses for Radiologists

Artificial intelligence (AI) systems play an increasingly important role in all parts of the imaging chain, from image creation to image interpretation to report generation. In order to responsibly manage radiology AI systems and make informed purchase decisions about them, radiologists must understand the underlying principles of AI. Our task force was formed by the Radiology Research Alliance (RRA) of the Association of University Radiologists to identify and summarize a curated list of current educational materials available for radiologists.

OTHR-06. PACS Lesion Tracking Tool provides real time automatic information on brain tumor metastasis growth curves and RECIST criteria

Objective

Communicating metastatic brain treatment response can be complicated. A widely used method to assess clinical response is called response evaluation criteria in solid tumors or RECIST. In our study, we use a PACS Lesion Tracking Tool (TT) to assess intracranial metastasis using RECIST criteria. We predict that the TT will be superior to the standard radiology reports.

Methods

Nuance ® mPowerTM was used to identify 30 patients with brain metastasis who received brain MRI from 4/2020–4/2021. Patient’s first brain MRI with metastasis was set as baseline and subsequent 3 brain MRI studies were examined. All lesions were measured on post-gadolinium sequence and defined as target lesions or new lesions. The TT was used to measure lesion size over time with creation of growth curves and RECIST outcomes, which include stable disease, progressive disease, partial response, or complete response. Subsequently, RECIST evaluations were compared with radiologic impressions for discrepancy, and further evaluations were made to see if it made a clinical difference in patient management and/or provide additional useful information. These evaluations were given a rating of agree/yes, equivocal, or disagree/no. They were assessed by 3 neuroradiologists.

Results

Number of lesions ranged from 1–27. The assessments from 3 neuroradiologists were averaged. Comparing impression versus RECIST evaluation, the results demonstrated the following: 8/30 disagreement, 4/30 equivocal, and 18/30 agreement. Using more stringent criteria, assessing whether the TT would result in either change in patient management or provide additional useful information, the results were the following: 6/30 yes, 4/30 equivocal, and 20/30 no.

Discussion

In addition to providing real time RECIST criteria evaluations and visually descriptive lesion growth tables, the TT was easy to use. Interpretation of these additional data provided more clarity and was found to be superior to standard radiology report.

Systematic review and meta-analysis on the safety of dalbavancin

BACKGROUND

Dalbavancin is a semisynthetic lipoglycopeptide antimicrobial agent with activity against Gram-positive bacteria including anaerobes.

RESEARCH DESIGN AND METHODS

Meta-analysis of randomized control trials and large case series (more than 20 patients), were identified by searching Pubmed and Cochrane databases through 14 December 2020.

RESULTS

3,073 patients from 6 RCTs met the inclusion criteria. Treatment emergent adverse effects were described in 30.6% dalbavancin patients, and 38.1% patients with other treatments. Our meta-analysis supports favorable results for dalbavancin treatment (OR 0.79; 95%CI 0.66-0.94; p = 0.01). 2.74% dalbavancin patients had to discontinue treatment versus 2.49% patients on other antibiotics. 4.80% dalbavancin patients versus 5.30% patients with other treatments had severe adverse events. 0.31% in the dalbavancin group and 0.95% receiving other antibiotics died. There was no statistically significant difference in severe adverse effects with OR 0.77; 95% CI 0.52-1.14; p = 0.19. Dalbavancin therapy was shown to have statistically significant lower mortality rate (OR 0.26; 95% CI 0.07-0.90; p = 0.03). Observational studies reported few side effects but included a heterogeneous population of patients concerning their diagnosis and the duration of antibiotic treatment.

CONCLUSIONS

Dalbavancin has comparable safety profile relative to other antibiotics and is well-tolerated.

A Bibliometric Evaluation of the Top 100 Cited Dimethyl Fumarate Articles

Dimethyl fumarate is a cytoprotective and immunomodulatory drug used in the treatment of multiple sclerosis. We performed a bibliometric study examining the characteristics and trends of the top 100 cited articles that include dimethyl fumarate in the title. On 21 September 2020 we carried out an electronic search in the Web of Science (WOS), seeking articles that include the following terms within the title: dimethyl fumarate, BG-12, or Tecfidera. To focus our investigation on original research, we refined the search to include only articles, early access, others, case report, and clinical trials. We obtained a total of 1115 items, which were cited 7169 times, had a citation density of 6.43 citations/item, and an h-index of 40. Around 2010, there was a jump in the number of published articles per year, rising from 5 articles/year up to 12 articles/year. We sorted all the items by the number of citations and selected the top 100 most cited (T100). The T100 had 4164 citations, with a density of 37 citations/year and contained 16 classic research articles. They were published between 1961 and 2018; the years 2010-2018 amassed nearly 80% of the T100. We noted 17 research areas with articles in the T100. Of these, the number one ranking went to neurosciences/neurology with 39 articles, and chemistry ranked second on the T100 list with 14 items. We noticed that the percentage of articles belonging to different journals changed depending on the time period. Chemistry held the highest number of papers during 1961-2000, while pharmacology andneurosciences/neurology led the 2001-2018 interval. A total of 478 authors from 145 institutions and 25 countries were included in the T100 ranking. The paper by Gold R et al. was the most successful with 14 articles, 1.823 citations and a density of 140.23 citations/year. The biotechnological company Biogen led the T100 list with 20 articles. With 59 published articles, the USA was the leading country in publications. We concluded that this study analyzed the use of and research on dimethyl fumarate from a different perspective, which will allow the readership (expert or not) to understand the relevance of classic and recent literature on this topic.

A bibliometric evaluation of the top 100 cited natalizumab articles

Natalizumab is being used in recurrent multiple sclerosis despite its history of market withdrawal due to lethal cases. We have carried out a bibliometric analysis of this drug from 1999 to February 2020 in order to assess the real impact of the use natalizumab with the goal to identify the key articles that sustain the current knowledge on the therapeutic possibilities of this compound. We have extracted from the Web of Science the top 100 most cited records (T100) and tabulated data on the journal, authors, publication year, number of citations, countries and institutions of publication, T100-records, citation density and citations per record of the works. The 100 most cited articles were selected from a total of 32,507 citations out of 2817 publications with an h-number of 74, 11.54 citations/publication, and a density of 1544.79 citations/year. Citations ranged from 63 of the paper placed in the 100th position (T100) to 1940 of the paper in the first position (T1). T2 was cited 888 times, and the difference in the number of citations between T1 and T2 was higher than that between T2 and T10. T1, T2 and T3 are clinical trials. When articles are arranged by institution and nationality having more than 10 T100 articles, biotechnology company Biogen and the USA, respectively, lead the ranking, but we also find that 8 out of 10 are academic European institutions. A co-authorship analysis reveals an intense collaborative activity between countries and institutions. We conclude that the clinical and academic communities have shown a sustained interest in natalizumab for the therapy of recurrent multiple sclerosis over the last 20 years.

Erratum: Construction of a Machine Learning Dataset through Collaboration: The RSNA 2019 Brain CT Hemorrhage Challenge

[This corrects the article DOI: 10.1148/ryai.2020190211.].

Construction of a Machine Learning Dataset through Collaboration: The RSNA 2019 Brain CT Hemorrhage Challenge

This dataset is composed of annotations of the five hemorrhage subtypes (subarachnoid, intraventricular, subdural, epidural, and intraparenchymal hemorrhage) typically encountered at brain CT.

Paradoxical Reaction in a Patient with Co-Occurring Tuberculous Meningitis and Pott’s Disease

Patient: Male, 36

Final Diagnosis: TB paradoxical reaction

Symptoms: Back pain • diplopia • Headache

Medication: —

Clinical Procedure: —

Specialty: Infectious Diseases

Estimating patient dose from x-ray tube output metrics: automated measurement of patient size from CT images enables large-scale size-specific dose estimates

PURPOSE

To test the hypothesis that patient size can be accurately calculated from axial computed tomographic (CT) images, including correction for the effects of anatomy truncation that occur in routine clinical CT image reconstruction.

MATERIALS AND METHODS

Institutional review board approval was obtained for this HIPAA-compliant study, with waiver of informed consent. Water-equivalent diameter (D(W)) was computed from the attenuation-area product of each image within 50 adult CT scans of the thorax and of the abdomen and pelvis and was also measured for maximal field of view (FOV) reconstructions. Linear regression models were created to compare D(W) with the effective diameter (D(eff)) used to select size-specific volume CT dose index (CTDI(vol)) conversion factors as defined in report 204 of the American Association of Physicists in Medicine. Linear regression models relating reductions in measured D(W) to a metric of anatomy truncation were used to compensate for the effects of clinical image truncation.

RESULTS

In the thorax, D(W)versus D(eff) had an R(2) of 0.51 (n = 200, 50 patients at four anatomic locations); in the abdomen and pelvis, R(2) was 0.90 (n = 150, 50 patients at three anatomic locations). By correcting for image truncation, the proportion of clinically reconstructed images with an extracted D(W) within ±5% of the maximal FOV D(W) increased from 54% to 90% in the thorax (n = 3602 images) and from 95% to 100% in the abdomen and pelvis (6181 images).

CONCLUSION

The D(W) extracted from axial CT images is a reliable measure of patient size, and varying degrees of clinical image truncation can be readily corrected. Automated measurement of patient size combined with CT radiation exposure metrics may enable patient-specific dose estimation on a large scale.

Exposing exposure: automated anatomy-specific CT radiation exposure extraction for quality assurance and radiation monitoring

PURPOSE

To develop and validate an informatics toolkit that extracts anatomy-specific computed tomography (CT) radiation exposure metrics (volume CT dose index and dose-length product) from existing digital image archives through optical character recognition of CT dose report screen captures (dose screens) combined with Digital Imaging and Communications in Medicine attributes.

MATERIALS AND METHODS

This institutional review board-approved HIPAA-compliant study was performed in a large urban health care delivery network. Data were drawn from a random sample of CT encounters that occurred between 2000 and 2010; images from these encounters were contained within the enterprise image archive, which encompassed images obtained at an adult academic tertiary referral hospital and its affiliated sites, including a cancer center, a community hospital, and outpatient imaging centers, as well as images imported from other facilities. Software was validated by using 150 randomly selected encounters for each major CT scanner manufacturer, with outcome measures of dose screen retrieval rate (proportion of correctly located dose screens) and anatomic assignment precision (proportion of extracted exposure data with correctly assigned anatomic region, such as head, chest, or abdomen and pelvis). The 95% binomial confidence intervals (CIs) were calculated for discrete proportions, and CIs were derived from the standard error of the mean for continuous variables. After validation, the informatics toolkit was used to populate an exposure repository from a cohort of 54 549 CT encounters; of which 29 948 had available dose screens.

RESULTS

Validation yielded a dose screen retrieval rate of 99% (597 of 605 CT encounters; 95% CI: 98%, 100%) and an anatomic assignment precision of 94% (summed DLP fraction correct 563 in 600 CT encounters; 95% CI: 92%, 96%). Patient safety applications of the resulting data repository include benchmarking between institutions, CT protocol quality control and optimization, and cumulative patient- and anatomy-specific radiation exposure monitoring.

CONCLUSION

Large-scale anatomy-specific radiation exposure data repositories can be created with high fidelity from existing digital image archives by using open-source informatics tools.

Exposing exposure: enhancing patient safety through automated data mining of nuclear medicine reports for quality assurance and organ dose monitoring

PURPOSE

To develop and validate an open-source informatics toolkit capable of creating a radiation exposure data repository from existing nuclear medicine report archives and to demonstrate potential applications of such data for quality assurance and longitudinal patient-specific radiation dose monitoring.

MATERIALS AND METHODS

This study was institutional review board approved and HIPAA compliant. Informed consent was waived. An open-source toolkit designed to automate the extraction of data on radiopharmaceuticals and administered activities from nuclear medicine reports was developed. After iterative code training, manual validation was performed on 2359 nuclear medicine reports randomly selected from September 17, 1985, to February 28, 2011. Recall (sensitivity) and precision (positive predictive value) were calculated with 95% binomial confidence intervals. From the resultant institutional data repository, examples of usage in quality assurance efforts and patient-specific longitudinal radiation dose monitoring obtained by calculating organ doses from the administered activity and radiopharmaceutical of each examination were provided.

RESULTS

Validation statistics yielded a combined recall of 97.6% ± 0.7 (95% confidence interval) and precision of 98.7% ± 0.5. Histograms of administered activity for fluorine 18 fluorodeoxyglucose and iodine 131 sodium iodide were generated. An organ dose heatmap which displays a sample patient's dose accumulation from multiple nuclear medicine examinations was created.

CONCLUSION

Large-scale repositories of radiation exposure data can be extracted from institutional nuclear medicine report archives with high recall and precision. Such repositories enable new approaches in radiation exposure patient safety initiatives and patient-specific radiation dose monitoring.

Mitochondrial biology in Alzheimer's disease pathogenesis

Despite the increasing knowledge of Alzheimer's disease (AD) management with novel pharmacologic agents, most of them are only transiently fixing symptomatic pathology. Currently there is rapid growth in the field of neuroprotective pharmacology and increasing focus on the involvement of mitochondria in this devastating disease. This review is directed at understanding the role of mitochondria-mediated pathways in AD and integrating basic biology of the mitochondria with knowledge of possible pharmacologic targets for AD treatment in an attempt to elucidate novel mitochondria-driven therapeutic interventions useful to both clinical and basic research.

Gyrification patterns in monozygotic twin pairs varying in discordance for autism

In order to disentangle genetic and environmental contributions to cortical anomalies in children with autism, we investigated cortical folding patterns in a cohort of 14 monozygotic (MZ) twin pairs who displayed a range of phenotypic discordance for autism, and 14 typically developing community controls. Cortical folding was assessed with the gyrification index, which was calculated on high resolution anatomic MR images. We found that the cortical folding patterns across most lobar regions of the cerebral cortex was highly discordant within MZ twin pairs. In addition, children with autism and their co-twins exhibited increased cortical folding in the right parietal lobe, relative to age- and gender-matched typical developing children. Increased folding in the right parietal lobe was associated with more symptoms of autism for co-twins. Finally, the robust association between cortical folding and IQ observed in typical children was not observed in either children with autism or their co-twins. These findings, which contribute to our understanding of the limits of genetic liability in autism, suggest that anomalies in the structural integrity of the cortex in this PDD may disrupt the association between cortical folding and intelligence that has been reported in typical individuals, and may account, in part, for the deficits in visual spatial attention and in social cognition that have been reported in children with autism.

Neuroanatomic alterations and social and communication deficits in monozygotic twins discordant for autism disorder

OBJECTIVE

Investigating neuroanatomic differences in monozygotic twins who are discordant for autism can help unravel the relative contributions of genetics and environment to this pervasive developmental disorder. The authors used magnetic resonance imaging (MRI) to investigate several brain regions of interest in monozygotic twins who varied in degree of phenotypic discordance for narrowly defined autism.

METHOD

The subjects were 14 pairs of monozygotic twins between the ages of 5 and 14 years old and 14 singleton age- and gender-matched typically developing comparison subjects. The monozygotic twin group was a cohort of children with narrowly defined autistic deficits and their co-twins who presented with varying levels of autistic deficits. High-resolution MRIs were acquired and volumetric/area measurements obtained for the frontal lobe, amygdala, and hippocampus and subregions of the prefrontal cortex, corpus callosum, and cerebellar vermis.

RESULTS

No neurovolumetric/area differences were found between twin pairs. Relative to typically developing comparison subjects, dorsolateral prefrontal cortex volumes and anterior areas of the corpus callosum were significantly altered in autistic twins, and volumes of the posterior vermis were altered in both autistic twins and co-twins. Intraclass correlation analysis of brain volumes between children with autism and their co-twins indicated that the degree of within-pair neuroanatomic concordance varied with brain region. In the group of subjects with narrowly defined autism only, dorsolateral prefrontal cortex, amygdala, and posterior vermis volumes were significantly associated with the severity of autism based on scores from the Autism Diagnostic Observation Schedule-Generic.

CONCLUSIONS

These findings support previous research demonstrating alterations in the prefrontal cortex, corpus callosum, and posterior vermis in children with autism and further suggest that alterations are associated with the severity of the autism phenotype. Continued research involving twins who are concordant and discordant for autism is essential to disentangle the genetic and environmental contributions to autism.

Minocycline and Cytoprotection: Shedding New Light on a Shadowy Controversy

In this review we explore and integrate the knowledge of the plausible pharmacological targets that could explain the new application for the well known semi-synthetic, tetracycline-derivate minocycline as a cytoprotective drug. In doing so, we will analyze the possible mechanisms to elucidate the potential cytoprotective properties of minocycline. We address its anti-oxidant action ranging from its structure to its capacity to modulate the expression of oxidant-related enzymes such as nitric oxide synthase. The pharmacological targets responsible for its anti-inflammatory effects are surveyed. The effects of this antibiotic are making its marks on intracellular pathways related to neurodegenerative processes such as mitochondrially-mediated apoptosis, including minocycline-modulated effects on the expression of apoptotic proteins. Finally, we will explore the effects of minocycline on metalloproteinases, enzymes implicated in the modulation of cerebrovascular post-ischemic oxidative reperfusion injury, and new targets. In conclusion, we shed new light on the shadowy controversy of minocycline's potential cytoprotective mechanisms and targets of action.