Diagnostic Aid Research Articles

Development and Evaluation of a GPT4-Based Orofacial Pain Clinical Decision Support System

Background: Orofacial pain (OFP) encompasses a complex array of conditions affecting the face, mouth, and jaws, often leading to significant diagnostic challenges and high rates of misdiagnosis. Artificial intelligence, particularly large language models like GPT4 (OpenAI, San Francisco, CA, USA), offers potential as a diagnostic aid in healthcare settings. Objective: To evaluate the diagnostic accuracy of GPT4 in OFP cases as a clinical decision support system (CDSS) and compare its performance against treating clinicians, expert evaluators, medical students, and general practitioners. Methods: A total of 100 anonymized patient case descriptions involving diverse OFP conditions were collected. GPT4 was prompted to generate primary and differential diagnoses for each case using the International Classification of Orofacial Pain (ICOP) criteria. Diagnoses were compared to gold-standard diagnoses established by treating clinicians, and a scoring system was used to assess accuracy at three hierarchical ICOP levels. A subset of 24 cases was also evaluated by two clinical experts, two final-year medical students, and two general practitioners for comparative analysis. Diagnostic performance and interrater reliability were calculated. Results: GPT4 achieved the highest accuracy level (ICOP level 3) in 38% of cases, with an overall diagnostic performance score of 157 out of 300 points (52%). The model provided accurate differential diagnoses in 80% of cases (400 out of 500 points). In the subset of 24 cases, the model’s performance was comparable to non-expert human evaluators but was surpassed by clinical experts, who correctly diagnosed 54% of cases at level 3. GPT4 demonstrated high accuracy in specific categories, correctly diagnosing 81% of trigeminal neuralgia cases at level 3. Interrater reliability between GPT4 and human evaluators was low (κ = 0.219, p < 0.001), indicating variability in diagnostic agreement. Conclusions: GPT4 shows promise as a CDSS for OFP by improving diagnostic accuracy and offering structured differential diagnoses. While not yet outperforming expert clinicians, GPT4 can augment diagnostic workflows, particularly in primary care or educational settings. Effective integration into clinical practice requires adherence to rigorous guidelines, thorough validation, and ongoing professional oversight to ensure patient safety and diagnostic reliability.

Small cell carcinoma of the ovary of hypercalcaemic type: a clinicopathological analysis of sixteen cases

Objective: To investigate the clinicopathological, molecular pathological features, and family genetic pedigree of small cell carcinoma of the ovary, hypercalcemic type (SCCOHT). Methods: A total of 16 cases of SCCOHT diagnosed in Obstetrics and Gynecology Hospital of Fudan University from January 2013 to January 2023 were collected. The clinicopathologic features, SMARCA4/2/B1 protein expression, outcomes and SMARCA4 gene detection were reported. A follow-up study was also carried out. Results: The average age at diagnosis was 28.7 years (range 17-38 years). The preoperative calcium level was evaluated in 3 of 6 patients. The tumor was unilateral in all 16 cases, ranged from 8 to 26 cm (average 15.8 cm) in the greatest dimension. Extraovarian spread was present in 7 cases. In 10 cases, the tumors were initially misinterpreted as other ovarian neoplasms. BRG1 and BRM expression by immunohistochemistry were all lost in detected cases, while INI1 exhibited retained nuclear expression. All BRM-negative SCCOHTs also lacked BRG1 protein,but retained INI1 expression. SCCOHTs were only focally positive for EMA, CKpan, Calretinin, SALL4, and diffusely positive for WT1. Two of nine cases exhibited mutation-type p53 immunoreactivity. Ki-67 index was 58% on an average. ER, PR, FOXL2, α-inhibin, chromogranin A and LCA were negative in all the cases. SMARCA4 sequencing was available in 8 cases of SCCOHT, which revealed a germline SMARCA4 mutation in one patient, and others carried somatic mutation. Furthermore, two daughters, mother and an aunt of a patient with germline mutation were reported to be SMARCA4 mutation carriers. Follow-up was available for 15 patients, and the 6-month, 1-year and 2-year survival rate was 65.8%, 45.1%, and 22.6%, respectively. For patients in FIGO stages Ⅱ+Ⅲ, 6-month, 1-year survival rate was 53.6% and 35.7% respectively, compared to 80% (6-month) and 60% (1-year) in patients of staged I (P=0.358). Conclusions: With dismal prognosis of SCCOHT, accurate diagnosis is necessary. The typical age distribution, a panel of various staining results, especially concomitant loss of BRG1 and BRM may be of diagnostic aid and can be used to distinguish SCCOHT from its histological mimics. After the diagnosis of SCCOHT, genetic testing and genetic counseling are recommended.

O USO DA TERMOGRAFIA NO DIAGNÓSTICO DA SÍNDROME FIBROMIÁLGICA

Fibromyalgia is a complex rheumatic disease characterized by diffuse pain, fatigue, insomnia and can be associated with emotional problems such as depression and anxiety. There is no validated method for accurate diagnosis. Thermography makes it possible to identify points of high sensitivity with a painful reaction using special cameras coupled to various thermometers. The aim of this study was to characterize the use of thermography as a diagnostic aid for FM. This is a bibliographic review, using databases (Scielo, Google Scholar, Pubmed, Portal BVsalud, and BDTD). It included research from 2014 to 2024 that was not a literature review. The results show that thermography has high specificity for identifying hot spots, but does not have high sensitivity for identifying chronic pain, so it was not possible to verify the presence of pain, but rather vasomotor changes in all areas of painful projection, identified on palpation. We conclude that thermography is viable for complementing and monitoring the treatment of a patient with fibromyalgia associated with other conditions and clinical findings, but it should not be used in isolation.

Comprehensive Exploration of Biomedical Science in Transforming Global Health

Biomedical science has supervised tremendous changes in what the world considers health and how diseases are prevented, diagnosed, and treated. From the discovery of vaccines and diagnostic aids to innovations in genetics and genomic personalized medicine, spot the discipline as a key determinant of advancing the delivery and models of health Systems globally. This paper aims to review the effect of biomedical science on global health with a focus on retrospect and prospects of diseases, health, and crises. Thus, relying on up-to-date research, cases, and world health studies, the work reveals successes and failures. As for recommendations, they are to encourage innovation, extend the usage of biomedical technologies and involve individuals and disciplines expecting to make a positive change in the coming years that is stable and effective.Biomedical Science

Establishment of an MRI-based radiomics model for distinguishing between intramedullary spinal cord tumor and tumefactive demyelinating lesion

ObjectiveDifferentiating intramedullary spinal cord tumor (IMSCT) from spinal cord tumefactive demyelinating lesion (scTDL) remains challenging with standard diagnostic approaches. This study aims to develop and evaluate the effectiveness of a magnetic resonance imaging (MRI)-based radiomics model for distinguishing scTDL from IMSCT before treatment initiation.MethodsA total of 75 patients were analyzed in this retrospective study, comprising 55 with IMSCT and 20 with scTDL. Radiomics features were extracted from T1- and T2-weighted imaging (T1&T2WI) scans upon admission. Ten classification algorithms were employed: logistic regression (LR); naive bayes (NaiveBayes); support vector machine (SVM); k nearest neighbors (KNN); random forest (RF); extra trees (ExtraTrees); eXtreme gradient boosting (XGBoost); light gradient boosting machine (LightGBM); gradient boosting (GradientBoosting); and multi-Layer perceptron (MLP). The performance of the optimal model was then compared to radiologists' assessments.ResultsThis study developed 30 predictive models using ten classifiers across two imaging sequences. The MLP model with two sequences (T1&T2WI) emerged as the most effective one, showing superior accuracy in MRI analysis with an area under the curve (AUC) of 0.991 in training and 0.962 in testing. Moreover, statistical analyses highlighted the radiomics model significantly outperformed radiologists' assessments (p < 0.05) in distinguishing between IMSCT and scTDL.ConclusionWe present an MRI-based radiomics model with high diagnostic accuracy in differentiating IMSCT from scTDL. The model’s performance was comparable to junior radiologists, highlighting its potential as an effective diagnostic aid in clinical practice.

Cutaneous T cell lymphoma atlas reveals malignant TH2 cells supported by a B cell-rich tumor microenvironment

Cutaneous T cell lymphoma (CTCL) is a potentially fatal clonal malignancy of T cells primarily affecting the skin. The most common form of CTCL, mycosis fungoides, can be difficult to diagnose, resulting in treatment delay. We performed single-cell and spatial transcriptomics analysis of skin from patients with mycosis fungoides-type CTCL and an integrated comparative analysis with human skin cell atlas datasets from healthy and inflamed skin. We revealed the co-optation of T helper 2 (TH2) cell-immune gene programs by malignant CTCL cells and modeling of the tumor microenvironment to support their survival. We identified MHC-II+ fibroblasts and dendritic cells that can maintain TH2 cell-like tumor cells. CTCL tumor cells are spatially associated with B cells, forming tertiary lymphoid structure-like aggregates. Finally, we validated the enrichment of B cells in CTCL and its association with disease progression across three independent patient cohorts. Our findings provide diagnostic aids, potential biomarkers for disease staging and therapeutic strategies for CTCL.

Abstract B012: M-PACT: Methylation-based predictive algorithm for CNS tumor liquid biopsies

Abstract Background: DNA methylation-based classification has been transformative in the molecular diagnostics of pediatric central nervous system (CNS) tumors. Current diagnostic pipelines integrate histopathology and molecular tools in accordance with the WHO classification of CNS tumors which rely on the availability of tissue specimens. However, some tumors are not amenable to neurosurgical resection due to their high-risk locations and tissue collection is not routinely performed at relapse. In addition, molecular biomarkers are largely lacking for longitudinal disease monitoring in pediatric neuro-oncology. Liquid biopsies (LBs) have emerged as a minimally invasive, longitudinal source of tumor-derived cell-free DNA (cfDNA) ideally suited for detecting minimal residual disease (MRD). In addition, LBs provide a sensitive means of studying tumor evolution at high temporal resolution. Success rates of LB analyses in neuro-oncology have varied drastically between studies, warranting the deployment of more robust and reproducible assays. Methods: We applied a novel cfDNA methylation-based workflow to a large cohort of cerebrospinal fluid (CSF) samples collected from pediatric brain tumor patients (n&amp;gt;200 patients). Enzymatic methylation sequencing (EM-seq) was adapted for CSF-derived cfDNA inputs in the picogram range, enabling the acquisition of genome-wide cfDNA methylation profiles across the cohort. Utilizing non-oncological cfDNA profiles and CNS tumor DNA methylation datasets as a reference, we developed an innovative computational pipeline that incorporates DNA methylation imputation and deep learning of a neural network. In addition, deconvolution of cfDNA profiles facilitated classification of low tumor burden samples, including those with balanced genomes that would have been missed using previous generation assays. Results: Our CSF-based classifier, M-PACT (Methylation-based Predictive Algorithm for CNS Tumors), yielded accurate tumor detection and entity prediction (AUC=0.9) in our benchmarking cohort (n&amp;gt;100 CSF samples). Proof-of-concept studies showed the potential of this methodology to track tumor evolution in serial CSF samples at both the genetic and epigenetic level, illuminating potential mechanisms driving treatment resistance and recurrence. Conclusions: Collectively, this study provides the framework for robust methylation- based tumor detection and classification of CSF LBs. M-PACT can be applied to aid in tumor diagnostics and to detect MRD during follow-up, warranting prospective validation of its broad applicability and clinical utility in future trials. Citation Format: Tom T. Fischer, Kyle S. Smith, Katie Han, Anna Kostecka, Hong Lin, Daniel Senfter, Tatjana Wedig, Nathalie Schwarz, Natalia Stepien, Sibylle Madlener, Christine Haberler, Esther Hulleman, Sandeep K. Dhanda, Stefan M. Pfister, Santhosh Upadhyaya, Amar Gajjar, Giles W. Robinson, Joonas Haapasalo, Hannu Haapasalo, Kristiina Nordfors, Johannes Gojo, Kendra K. Maass, Kristian W. Pajtler, Paul A. Northcott. M-PACT: Methylation-based predictive algorithm for CNS tumor liquid biopsies [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr B012.

Preoperative Prediction of Perineural Invasion in Pancreatic Ductal Adenocarcinoma Using Machine Learning Radiomics Based on Contrast-Enhanced CT Imaging.

The objective of the study is to assess the clinical value of machine learning radiomics based on contrast-enhanced computed tomography (CECT) images in preoperative prediction of perineural invasion (PNI) status in pancreatic ductal adenocarcinoma (PDAC). A total of 143 patients with PDAC were enrolled in this retrospective study (training group, n = 100; test group, n = 43). Radiomics features were extracted from CECT images and selected by the Mann-Whitney U-test, Pearson correlation coefficient, and least absolute shrinkage and selection operator (LASSO). The logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), and decision tree (DT) algorithms were trained to build radiomics models by radiomic features. Multivariate logistic regression was employed to identify independent predictors and establish clinical models. A combined model was constructed by integrating clinical and radiomics features. Model performances were assessed by receiver operating characteristic curves (ROCs) and decision curve analyses (DCAs). A total of 788 radiomics features were extracted from CECT images, of which 14 were identified as significant through the three-step selection process. Among the machine learning models, the SVM radiomics model exhibited the highest predictive performance in the test group, achieving an area under the curve (AUC) of 0.831, accuracy of 0.698, sensitivity of 0.677, and specificity of 0.750. After logistic regression screening, the clinical model was established using carbohydrate antigen 19-9 (CA199) levels as one independent predictor. In the test group, the clinical model demonstrated an AUC of 0.644, accuracy of 0.744, sensitivity of 0.871, and specificity of 0.417. The combined model showed improved performance compared to both the clinical and radiomics models in the test group, with an AUC of 0.844, accuracy of 0.767, sensitivity of 0.806, and specificity of 0.667. Subsequently, DCA of the combined model indicated optimal clinical value for predicting PNI status. Machine learning radiomics models can accurately predict PNI status in patients with pancreatic ductal adenocarcinoma. The combined model, which incorporates clinical and radiomics features, enhances preoperative diagnostic performance and aids in the selection of treatment methods.

Utilization of Single-Pulse Transcranial-Evoked Potentials in Neurological and Psychiatric Clinical Practice: A Narrative Review.

Background: The utility of single-pulse TMS (transcranial magnetic stimulation)-evoked EEG (electroencephalograph) potentials (TEPs) has been extensively studied in the past three decades. TEPs have been shown to provide insights into features of cortical excitability and connectivity, reflecting mechanisms of excitatory/inhibitory balance, in various neurological and psychiatric conditions. In the present study, we sought to review and summarize the most studied neurological and psychiatric clinical indications utilizing single-pulse TEP and describe its promise as an informative novel tool for the evaluation of brain physiology. Methods: A thorough search of PubMed, Embase, and Google Scholar for original research utilizing single-pulse TMS-EEG and the measurement of TEP was conducted. Our review focused on the indications and outcomes most clinically relevant, commonly studied, and well-supported scientifically. Results: We included a total of 55 publications and summarized them by clinical application. We categorized these publications into seven sub-sections: healthy aging, Alzheimer's disease (AD), disorders of consciousness (DOCs), stroke rehabilitation and recovery, major depressive disorder (MDD), Parkinson's disease (PD), as well as prediction and monitoring of treatment response. Conclusions: TEP is a useful measurement of mechanisms underlying neuronal networks. It may be utilized in several clinical applications. Its most prominent uses include monitoring of consciousness levels in DOCs, monitoring and prediction of treatment response in MDD, and diagnosis of AD. Additional applications including the monitoring of stroke rehabilitation and recovery, as well as a diagnostic aid for PD, have also shown encouraging results but require further evidence from randomized controlled trials (RCTs).

DNA methylation classifier to diagnose pancreatic ductal adenocarcinoma metastases from different anatomical sites

BackgroundWe have recently constructed a DNA methylation classifier that can discriminate between pancreatic ductal adenocarcinoma (PAAD) liver metastasis and intrahepatic cholangiocarcinoma (iCCA) with high accuracy (PAAD-iCCA-Classifier). PAAD is one of the leading causes of cancer of unknown primary and diagnosis is based on exclusion of other malignancies. Therefore, our focus was to investigate whether the PAAD-iCCA-Classifier can be used to diagnose PAAD metastases from other sites.MethodsFor this scope, the anomaly detection filter of the initial classifier was expanded by 8 additional mimicker carcinomas, amounting to a total of 10 carcinomas in the negative class. We validated the updated version of the classifier on a validation set, which consisted of a biological cohort (n = 3579) and a technical one (n = 15). We then assessed the performance of the classifier on a test set, which included a positive control cohort of 16 PAAD metastases from various sites and a cohort of 124 negative control samples consisting of 96 breast cancer metastases from 18 anatomical sites and 28 carcinoma metastases to the brain.ResultsThe updated PAAD-iCCA-Classifier achieved 98.21% accuracy on the biological validation samples, and on the technical validation ones it reached 100%. The classifier also correctly identified 15/16 (93.75%) metastases of the positive control as PAAD, and on the negative control, it correctly classified 122/124 samples (98.39%) for a 97.85% overall accuracy on the test set. We used this DNA methylation dataset to explore the organotropism of PAAD metastases and observed that PAAD liver metastases are distinct from PAAD peritoneal carcinomatosis and primary PAAD, and are characterized by specific copy number alterations and hypomethylation of enhancers involved in epithelial-mesenchymal-transition.ConclusionsThe updated PAAD-iCCA-Classifier (available at https://classifier.tgc-research.de/) can accurately classify PAAD samples from various metastatic sites and it can serve as a diagnostic aid.

Assessing the Utility of ColabFold and AlphaMissense in Determining Missense Variant Pathogenicity for Congenital Myasthenic Syndromes.

Congenital myasthenic syndromes (CMSs) are caused by variants in >30 genes with increasing numbers of variants of unknown significance (VUS) discovered by next-generation sequencing. Establishing VUS pathogenicity requires in vitro studies that slow diagnosis and treatment initiation. The recently developed protein structure prediction software AlphaFold2/ColabFold has revolutionized structural biology; such predictions have also been leveraged in AlphaMissense, which predicts ClinVar variant pathogenicity with 90% accuracy. Few reports, however, have tested these tools on rigorously characterized clinical data. We therefore assessed ColabFold and AlphaMissense as diagnostic aids for CMSs, using variants of the CHRN genes that encode the nicotinic acetylcholine receptor (nAChR). Utilizing a dataset of 61 clinically validated CHRN variants, (1) we evaluated the possibility of a ColabFold metric (either predicted structural disruption, prediction confidence, or prediction quality) that distinguishes variant pathogenicity; (2) we assessed AlphaMissense's ability to differentiate variant pathogenicity; and (3) we compared AlphaMissense to the existing pathogenicity prediction programs AlamutVP and EVE. Analyzing the variant effects on ColabFold CHRN structure prediction, prediction confidence, and prediction quality did not yield any reliable pathogenicity indicative metric. However, AlphaMissense predicted variant pathogenicity with 63.93% accuracy in our dataset-a much greater proportion than AlamutVP (27.87%) and EVE (28.33%). Emerging in silico tools can revolutionize genetic disease diagnosis-however, improvement, refinement, and clinical validation are imperative prior to practical acquisition.

Artificial Intelligence Efficacy as a Function of Trainee Interpreter Proficiency: Lessons from a Randomized Controlled Trial.

Recently, artificial intelligence tools have been deployed with increasing speed in educational and clinical settings. However, the use of artificial intelligence by trainees across different levels of experience has not been well-studied. This study investigates the impact of artificial intelligence assistance on the diagnostic accuracy for intracranial hemorrhage and large-vessel occlusion by medical students and resident trainees. This prospective study was conducted between March 2023 and October 2023. Medical students and resident trainees were asked to identify intracranial hemorrhage and large-vessel occlusion in 100 noncontrast head CTs and 100 head CTAs, respectively. One group received diagnostic aid simulating artificial intelligence for intracranial hemorrhage only (n = 26); the other, for large-vessel occlusion only (n = 28). Primary outcomes included accuracy, sensitivity, and specificity for intracranial hemorrhage/large-vessel occlusion detection without and with aid. Study interpretation time was a secondary outcome. Individual responses were pooled and analyzed with the t test; differences in continuous variables were assessed with ANOVA. Forty-eight participants completed the study, generating 10,779 intracranial hemorrhage or large-vessel occlusion interpretations. With diagnostic aid, medical student accuracy improved 11.0 points (P < .001) and resident trainee accuracy showed no significant change. Intracranial hemorrhage interpretation time increased with diagnostic aid for both groups (P < .001), while large-vessel occlusion interpretation time decreased for medical students (P < .001). Despite worse performance in the detection of the smallest-versus-largest hemorrhages at baseline, medical students were not more likely to accept a true-positive artificial intelligence result for these more difficult tasks. Both groups were considerably less accurate when disagreeing with the artificial intelligence or when supplied with an incorrect artificial intelligence result. This study demonstrated greater improvement in diagnostic accuracy with artificial intelligence for medical students compared with resident trainees. However, medical students were less likely than resident trainees to overrule incorrect artificial intelligence interpretations and were less accurate, even with diagnostic aid, than the artificial intelligence was by itself.

Intracranial pleomorphic liposarcoma misclassified as a pleomorphic xanthoastrocytoma by a DNA methylation classifier: illustrative case.

Recently, it has been shown that DNA methylation arrays and German Cancer Research Center (Deutsches Krebsforschungszentrum) methylation classifiers are useful aids in brain tumor diagnosis for cases in which histopathological diagnosis is difficult. However, not enough is known about diagnostic aids for intracranial liposarcoma (LPS). An 18-year-old woman with a history of natural killer/T-cell lymphoma, which had been treated with a bone marrow transplant and total body irradiation at age 11 years, presented with diplopia. Magnetic resonance imaging revealed a brain tumor in the posterior left temporal lobe, which was removed by craniotomy. The tumor was initially diagnosed as pleomorphic xanthoastrocytoma through histopathological and DNA methylation examination. She also had a soft tissue tumor in her left thigh, which was removed. It contained spindle cells with oval nuclei and highly pleomorphic cells and was diagnosed as radiation-induced LPS. Histopathological re-examination of the brain tumor led to a final diagnosis of pleomorphic LPS. In this report, the authors describe the case of a patient with an intracranial pleomorphic LPS that was initially classified as a pleomorphic xanthoastrocytoma by a DNA methylation classifier. Although DNA methylation classifiers are useful as diagnostic aids in cases in which definitive pathology is difficult to determine, there is a risk of misdiagnosis in some types of tumors. https://thejns.org/doi/10.3171/CASE24465.

Letter to Editor Regarding “Use of Artificial Intelligence Software to Detect Intracranial Aneurysms: A Comprehensive Stroke Center Experience”

Artificial intelligence (AI) is increasingly significant in neurosurgery, enhancing differential diagnosis, preoperative evaluation, and surgical precision. A recent study in World Neurosurgery evaluated AI's role in aneurysm detection, comparing conventional computed tomography angiography images with AI analysis. AI identified 33 potential aneurysms, with 16 confirmed by radiologists, demonstrating a sensitivity of 36%, specificity of 97.6%, and a negative predictive value of 96.2%. Although AI shows promise, it is recommended as a complementary tool rather than a replacement for standard care. Further studies highlight AI's role in detecting and managing intracranial aneurysms, preventing severe complications like subarachnoid hemorrhage. AI improves medical image analysis, reducing missed detections and false positives. A review of 47 AI-based studies noted that these algorithms enhance diagnostic accuracy and aid in precise aneurysm identification. Another study showed AI's superior precision in measuring intracranial aneurysm size compared to junior raters. A systematic review of studies from 2015 to 2023 confirmed AI's efficacy in detecting cerebral aneurysms, with high sensitivity for larger aneurysms. However, sensitivity for smaller aneurysms remained lower, indicating the need for further research. In conclusion, AI integration in aneurysm detection and management enhances diagnostic accuracy and precision. While current AI technologies show significant strengths, ongoing research is essential to address limitations and fully realize AI's potential in neurosurgery.

Prevalence and Significance of AGR2 Expression in Human Cancer.

Anterior gradient 2 (AGR2) is a resident endoplasmic reticulum (ER) protein with a vital role in embryonal development, mucus maturation, tissue regeneration, and wound healing. To determine the prevalence and clinical significance of AGR2 expression in cancer, a tissue microarray containing 14,966 tumors from 134 different tumor types and subtypes as well as 608 samples of 76 different normal tissue types was analyzed by immunohistochemistry (IHC). AGR2 positivity was found in 103 of 134 tumor categories, and 83 tumor categories contained at least one strongly positive case. AGR2 expression was most frequently seen in tumors of the female genital tract, particularly adenocarcinomas (up to 100%), various breast cancer subtypes (57.1%-100%), urothelial carcinoma (74.6%-100%), adenocarcinomas of the upper and lower gastrointestinal tract (93.6%-99.6%), and pancreaticobiliary cancers (65.2%-98.2%). AGR2 positivity was slightly less common in squamous cell carcinomas (46.4%-77.3%) and mainly absent in mesenchymal and lymphoid tumors. While AGR2 expression was only weak or absent in the normal thyroid, it was moderate to strong in 46.0% of adenomas, 52.8% of follicular carcinomas, and 81.8% of papillary carcinomas of the thyroid. High AGR2 expression was strongly linked to poor ISUP (p < 0.0001), Fuhrman (p < 0.0001), and Thoenes (p < 0.0001) grades as well as advanced pT stage (p = 0.0035) in clear cell renal cell carcinoma (ccRCC). Low AGR2 expression was associated with high BRE grade in breast cancer (p = 0.0049), nodal metastasis (p = 0.0275) and RAS mutation (p = 0.0136) in colorectal cancer, nodal metastasis (p = 0.0482) in endometrioid endometrial carcinoma, high grade in noninvasive urothelial carcinoma (p = 0.0003), and invasive tumor growth in urothelial carcinoma (p < 0.0001). It is concluded that AGR2 expression occurs in a broad range of different tumor entities and that AGR2 assessment may serve as a diagnostic aid for the distinction of thyroidal neoplasms and as a prognostic marker in various cancer types.

A Rare Case of the Bladder Endometriosis and Overview of the Literature

Endometriosis is a chronic, non-cancerous condition characterized by the growth of endometrial tissue outside the uterus. Isolated involvement of the bladder is rare and often part of a broader deep infiltrating pelvic endometriosis. Bladder endometriosis should be considered in cases of unexplained dysuria and urinary symptoms not clarified by imaging or pelvic examination. Magnetic resonance imaging can be used as a diagnostic aid. This case report presents two patients diagnosed with bladder endometriosis postoperatively. Conservative treatment with gestagen preparations was administered. After 3-6 months of follow-up, ultrasonography and symptom assessment showed no recurrence. Clinicians should be aware of bladder endometriosis as a potential cause of persistent urinary symptoms, given its rarity and often delayed diagnosis.

Large Language Model Influence on Diagnostic Reasoning

Large language models (LLMs) have shown promise in their performance on both multiple-choice and open-ended medical reasoning examinations, but it remains unknown whether the use of such tools improves physician diagnostic reasoning. To assess the effect of an LLM on physicians' diagnostic reasoning compared with conventional resources. A single-blind randomized clinical trial was conducted from November 29 to December 29, 2023. Using remote video conferencing and in-person participation across multiple academic medical institutions, physicians with training in family medicine, internal medicine, or emergency medicine were recruited. Participants were randomized to either access the LLM in addition to conventional diagnostic resources or conventional resources only, stratified by career stage. Participants were allocated 60 minutes to review up to 6 clinical vignettes. The primary outcome was performance on a standardized rubric of diagnostic performance based on differential diagnosis accuracy, appropriateness of supporting and opposing factors, and next diagnostic evaluation steps, validated and graded via blinded expert consensus. Secondary outcomes included time spent per case (in seconds) and final diagnosis accuracy. All analyses followed the intention-to-treat principle. A secondary exploratory analysis evaluated the standalone performance of the LLM by comparing the primary outcomes between the LLM alone group and the conventional resource group. Fifty physicians (26 attendings, 24 residents; median years in practice, 3 [IQR, 2-8]) participated virtually as well as at 1 in-person site. The median diagnostic reasoning score per case was 76% (IQR, 66%-87%) for the LLM group and 74% (IQR, 63%-84%) for the conventional resources-only group, with an adjusted difference of 2 percentage points (95% CI, -4 to 8 percentage points; P = .60). The median time spent per case for the LLM group was 519 (IQR, 371-668) seconds, compared with 565 (IQR, 456-788) seconds for the conventional resources group, with a time difference of -82 (95% CI, -195 to 31; P = .20) seconds. The LLM alone scored 16 percentage points (95% CI, 2-30 percentage points; P = .03) higher than the conventional resources group. In this trial, the availability of an LLM to physicians as a diagnostic aid did not significantly improve clinical reasoning compared with conventional resources. The LLM alone demonstrated higher performance than both physician groups, indicating the need for technology and workforce development to realize the potential of physician-artificial intelligence collaboration in clinical practice. ClinicalTrials.gov Identifier: NCT06157944.

Assessing the impact of implantable loop recorder placement on anxiety level

Abstract Background Anxiety is a prolonged state of apprehension or worry that can be an independent predictor of cardiovascular events, associated with a 48% increased risk of cardiac death. About one-third of patients with arrhythmias present with high anxiety levels, and subsequently worse quality of life. In our experience of patient care, many patients feel anxious about the possibility of an arrhythmia that could lead to their demise, especially if their symptoms are undiagnosed. Purpose Our project aims to explore the change in patients’ anxiety levels following implantable loop recorder (ILR) placement. Methods This is a survey-based observational study that was performed in an outpatient clinic and hospital using validated General Anxiety Disorder-7 (GAD-7) questionnaires. Individuals who met indications to undergo ILR implantation including palpitations, syncope, subclinical atrial fibrillation, and cryptogenic stroke underwent baseline assessment of anxiety prior to implantation. Individuals had repeat assessments at 6 weeks and 12 weeks post-implantation to evaluate the effect of ILR placement. Results Our study included 50 individuals. Out of the initial subjects, 10/25 females and 11/25 males were lost to follow-up. The average initial GAD-7 score was 3.3 (Range 0, 15), average 6-week score was 1.6 (Range 0, 14), and average 12-week score was 0.9 (Range 0, 10). Total decrease in GAD-7 score after 6 weeks of having ILR was 50%. Total decrease in GAD-7 score after 12 weeks of ILR was statistically significant at 73% (P=0.03). Females showed greater improvement of GAD-7 scores in comparison to males at 6 weeks (67% vs 36%) and 12 weeks (81% vs 66%). Conclusion Studies have suggested an overall decreased level of anxiety in patients receiving implantable devices that have the potential to treat arrhythmias such as implantable cardiac defibrillators and pacemakers. Our results suggest that the benefit may extend to implantable devices that seek to provide further diagnostic information and aid in the prognostication of disease burden. As seen in our study population, GAD-7 scores showed dramatic improvement with greater percentage in female than male population. After the placement of loop recorders, many patients experience a sense of security with having a device that monitors them. Our study is the first to address the association between ILR and anxiety. Future studies evaluating this association would help increase the power of the study.Figure 1:Bar graph of GAD-7 Scores

Feature Extraction and Identification of Rheumatoid Nodules Using Advanced Image Processing Techniques

Background/Objectives: Accurate detection and classification of nodules in medical images, particularly rheumatoid nodules, are critical due to the varying nature of these nodules, where their specific type is often unknown before analysis. This study addresses the challenges of multi-class prediction in nodule detection, with a specific focus on rheumatoid nodules, by employing a comprehensive approach to feature extraction and classification. We utilized a diverse dataset of nodules, including rheumatoid nodules sourced from the DermNet dataset and local rheumatologists. Method: This study integrates 62 features, combining traditional image characteristics with advanced graph-based features derived from a superpixel graph constructed through Delaunay triangulation. The key steps include image preprocessing with anisotropic diffusion and Retinex enhancement, superpixel segmentation using SLIC, and graph-based feature extraction. Texture analysis was performed using Gray-Level Co-occurrence Matrix (GLCM) metrics, while shape analysis was conducted with Fourier descriptors. Vascular pattern recognition, crucial for identifying rheumatoid nodules, was enhanced using the Frangi filter. A Hybrid CNN–Transformer model was employed for feature fusion, and feature selection and hyperparameter tuning were optimized using Gray Wolf Optimization (GWO) and Particle Swarm Optimization (PSO). Feature importance was assessed using SHAP values. Results: The proposed methodology achieved an accuracy of 85%, with a precision of 0.85, a recall of 0.89, and an F1 measure of 0.87, demonstrating the effectiveness of the approach in detecting and classifying rheumatoid nodules in both binary and multi-class classification scenarios. Conclusions: This study presents a robust tool for the detection and classification of nodules, particularly rheumatoid nodules, in medical imaging, offering significant potential for improving diagnostic accuracy and aiding in the early identification of rheumatoid conditions.

Deep Learning Models for Anatomical Location Classification in Esophagogastroduodenoscopy Images and Videos: A Quantitative Evaluation with Clinical Data.

During gastroscopy, accurately identifying the anatomical locations of the gastrointestinal tract is crucial for developing diagnostic aids, such as lesion localization and blind spot alerts. This study utilized a dataset of 31,403 still images from 1000 patients with normal findings to annotate the anatomical locations within the images and develop a classification model. The model was then applied to videos of 20 esophagogastroduodenoscopy procedures, where it was validated for real-time location prediction. To address instability of predictions caused by independent frame-by-frame assessment, we implemented a hard-voting-based post-processing algorithm that aggregates results from seven consecutive frames, improving the overall accuracy. Among the tested models, InceptionV3 demonstrated superior performance for still images, achieving an F1 score of 79.79%, precision of 80.57%, and recall of 80.08%. For video data, the InceptionResNetV2 model performed best, achieving an F1 score of 61.37%, precision of 73.08%, and recall of 57.21%. These results indicate that the deep learning models not only achieved high accuracy in position recognition for still images but also performed well on video data. Additionally, the post-processing algorithm effectively stabilized the predictions, highlighting its potential for real-time endoscopic applications. This study demonstrates the feasibility of predicting the gastrointestinal tract locations during gastroscopy and suggests a promising path for the development of advanced diagnostic aids to assist clinicians. Furthermore, the location information generated by this model can be leveraged in future technologies, such as automated report generation and supporting follow-up examinations for patients.