Citations

A Multi-Center Comparison of Surgical Techniques for Corpus Callosotomy in Pediatric Drug-resistant Epilepsy Hansen D, Shandley S, Olaya J, Hauptman J, Auguste K, Ostendorf AP, Depositario-Cabacar DF, Wong-Kisiel LC, Reddy SB, McCormack MJ, Gonzalez-Giraldo E, Sullivan J, Pradeep J, Singh RK, Romanowski EF, McNamara NA, Ciliberto MA, Tatachar P, Shrey DW, Karakas C, Karia S, Kheder A, Gedela S, Alexander A, Eschbach K, Bolton J, Marashly A, Wolf S, McGoldrick P, Nangia S, Grinspan Z, Coryell J, Samanta D, Armstrong D, Perry MS. Epilepsia . 2023 Dec 7. Epub ahead of print. PMID: 38062633 . doi: 10.1111/epi.17853 Objectives: Corpus callosotomy (CC) is used to reduce seizures, primarily in patients with generalized drug-resistant epilepsy (DRE). The invasive nature of the procedure contributes to underutilization despite its potential superiority to other palliative procedures. The goal of this study was to use a multi-institutional epilepsy surgery database to characterize the use of CC across participating centers. Methods: Data were acquired from the Pediatric Epilepsy Research Consortium (PERC) Surgery Database, a prospective observational study collecting data on children 0-18 years referred for surgical evaluation of DRE across 22 U.S. pediatric epilepsy centers. Patient, epilepsy, and surgical characteristics were collected across multiple CC modalities. Outcomes and complications were recorded and analyzed statistically. Results: Eighty-three patients undergoing 85 CC procedures at 14 participating epilepsy centers met inclusion criteria. Mean age at seizure onset was 2.3 years (0-9.4); mean age for Phase I evaluation and surgical intervention were 9.45 (.1-20) and 10.46 (.2-20.6) years, respectively. Generalized seizure types were the most common (59%). Complete CC was performed in 88%. The majority of CC procedures (57%) were via open craniotomy, followed by laser interstitial thermal therapy (LiTT) (20%) and mini-craniotomy/endoscopic (mc/e) (22%). Mean operative times were significantly longer for LiTT, whereas mean estimated blood loss was greater in open cases. Complications occurred in 11 cases (13%) and differed significantly between surgical techniques (p < .001). There was no statistically significant difference in length of postoperative stay across approaches. Mean follow-up was 12.8 months (range 1-39). Favorable Engel outcomes were experienced by 37 (78.7%) of the patients who underwent craniotomy, 10 (58.8%) with LiTT, and 12 (63.2%) with mc/e; these differences were not statistically significant. Significance: CC is an effective surgical modality for children with DRE. Regardless of surgical modality, complication rates are acceptable and seizure outcomes generally favorable. Newer, less-invasive, surgical approaches may lead to increased adoption of this efficacious therapeutic option for pediatric DRE.

Background This study investigated the efficacy of ChatGPT-3.5 and ChatGPT-4 in assessing drug safety for patients with kidney diseases, comparing their performance to Micromedex, a well-established drug information source. Despite the perception of non-prescription medications and supplements as safe, risks exist, especially for those with kidney issues. The study's goal was to evaluate ChatGPT's versions for their potential in clinical decision-making regarding kidney disease patients. Method The research involved analyzing 124 common non-prescription medications and supplements using ChatGPT-3.5 and ChatGPT-4 with queries about their safety for people with kidney disease. The AI responses were categorized as “generally safe,” “potentially harmful,” or “unknown toxicity.” Simultaneously, these medications and supplements were assessed in Micromedex using similar categories, allowing for a comparison of the concordance between the two resources. Results Micromedex identified 85 (68.5%) medications as generally safe, 35 (28.2%) as potentially harmful, and 4 (3.2%) of unknown toxicity. ChatGPT-3.5 identified 89 (71.8%) as generally safe, 11 (8.9%) as potentially harmful, and 24 (19.3%) of unknown toxicity. GPT-4 identified 82 (66.1%) as generally safe, 29 (23.4%) as potentially harmful, and 13 (10.5%) of unknown toxicity. The overall agreement between Micromedex and ChatGPT-3.5 was 64.5% and ChatGPT-4 demonstrated a higher agreement at 81.4%. Notably, ChatGPT-3.5's suboptimal performance was primarily influenced by a lower concordance rate among supplements, standing at 60.3%. This discrepancy could be attributed to the limited data on supplements within ChatGPT-3.5, with supplements constituting 80% of medications identified as unknown. Conclusion ChatGPT's capabilities in evaluating the safety of non-prescription drugs and supplements for kidney disease patients are modest compared to established drug information resources. Neither ChatGPT-3.5 nor ChatGPT-4 can be currently recommended as reliable drug information sources for this demographic. The results highlight the need for further improvements in the model's accuracy and reliability in the medical domain.

AbstractBackground and AimEffective clinical event classification is essential for clinical research and quality improvement. The validation of artificial intelligence (AI) models like Generative Pre‐trained Transformer 4 (GPT‐4) for this task and comparison with conventional methods remains unexplored.MethodsWe evaluated the performance of the GPT‐4 model for classifying gastrointestinal (GI) bleeding episodes from 200 medical discharge summaries and compared the results with human review and an International Classification of Diseases (ICD) code‐based system. The analysis included accuracy, sensitivity, and specificity evaluation, using ground truth determined by physician reviewers.ResultsGPT‐4 exhibited an accuracy of 94.4% in identifying GI bleeding occurrences, outperforming ICD codes (accuracy 63.5%, P < 0.001). GPT‐4's accuracy was either slightly lower or statistically similar to individual human reviewers (Reviewer 1: 98.5%, P < 0.001; Reviewer 2: 90.8%, P = 0.170). For location classification, GPT‐4 achieved accuracies of 81.7% and 83.5% for confirmed and probable GI bleeding locations, respectively, with figures that were either slightly lower or comparable with those of human reviewers. GPT‐4 was highly efficient, analyzing the dataset in 12.7 min at a cost of 21.2 USD, whereas human reviewers required 8–9 h each.ConclusionOur study indicates GPT‐4 offers a reliable, cost‐efficient, and faster alternative to current clinical event classification methods, outperforming the conventional ICD coding system and performing comparably to individual expert human reviewers. Its implementation could facilitate more accurate and granular clinical research and quality audits. Future research should explore scalability, prompt and model tuning, and ethical implications of high‐performance AI models in clinical data processing.

AbstractALG1‐CDG is a rare, clinically variable metabolic disease, caused by the defect of adding the first mannose (Man) to N‐acetylglucosamine (GlcNAc2)‐pyrophosphate (PP)‐dolichol to the growing oligosaccharide chain, resulting in impaired N‐glycosylation of proteins. N‐glycosylation has a key role in functionality, stability, and half‐life of most proteins. Therefore, congenital defects of glycosylation typically are multisystem disorders. Here we report a 3‐year‐old patient with severe neurological, cardiovascular, respiratory, musculoskeletal and gastrointestinal symptoms. ALG1‐CDG was suggested based on exome sequencing and Western blot analysis. Despite her severe clinical manifestations and genetic diagnosis, serum transferrin glycoform analysis was normal. Western blot analysis of highly glycosylated proteins in fibroblasts revealed decreased intercellular adhesion molecule 1 (ICAM1), but normal lysosomal associated membrane protein 1 and 2 (LAMP1 and LAMP2) expression levels. Glycoproteomics in fibroblasts showed the presence of the abnormal tetrasacharide. Reviewing the literature, we found 86 reported ALG1‐CDG patients, but only one with normal transferrin analysis. Based on our results we would like to highlight the importance of multiple approaches in diagnosing ALG1‐CDG, as normal serum transferrin glycosylation or other biomarkers with normal expression levels can occur.

Abstract Coronary artery disease (CAD) is a leading health challenge worldwide. Exercise stress testing is a foundational non-invasive diagnostic tool. Nonetheless, its variable accuracy prompts the exploration of more reliable methods. Recent advancements in machine learning (ML), including deep learning and natural language processing, have shown potential in refining the interpretation of stress testing data. Adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we conducted a systematic review of ML applications in stress electrocardiogram (ECG) and stress echocardiography for CAD prognosis. Medical Literature Analysis and Retrieval System Online, Web of Science, and the Cochrane Library were used as databases. We analysed the ML models, outcomes, and performance metrics. Overall, seven relevant studies were identified. Machine-learning applications in stress ECGs resulted in sensitivity and specificity improvements. Some models achieved rates of above 96% in both metrics and reduced false positives by up to 21%. In stress echocardiography, ML models demonstrated an increase in diagnostic precision. Some models achieved specificity and sensitivity rates of up to 92.7 and 84.4%, respectively. Natural language processing applications enabled the categorization of stress echocardiography reports, with accuracy rates nearing 98%. Limitations include a small, retrospective study pool and the exclusion of nuclear stress testing, due to its well-documented status. This review indicates the potential of artificial intelligence applications in refining CAD stress testing assessment. Further development for real-world use is warranted.

The NCCN Guidelines for Prostate Cancer include recommendations for staging and risk assessment after a prostate cancer diagnosis and for the care of patients with localized, regional, recurrent, and metastatic disease. These NCCN Guidelines Insights summarize the panel's discussions for the 2024 update to the guidelines with regard to initial risk stratification, initial management of very-low-risk disease, and the treatment of nonmetastatic recurrence.

Abstract Context The risk of gestational diabetes mellitus (GDM) in twin pregnancies is more than double that of singleton pregnancies. Although twin pregnancies present unique challenges for fetal growth and prenatal management, the approach to GDM diagnosis and treatment is the same regardless of plurality. Data on pregnancy outcomes for individuals with GDM and a twin pregnancy are limited and conflicting. Objective To describe the maternal characteristics associated with GDM in twin pregnancies and to assess the associated pregnancy outcomes compared to twin pregnancies unaffected by GDM. Methods A retrospective cohort study was conducted at Mayo Clinic, Rochester, Minnesota, USA, and included predominantly Causasian women aged 18 to 45 years who received prenatal care for a twin pregnancy from 2017-2022. Maternal characteristics and a broad spectrum of pregnancy outcomes were evaluated. Universal GDM screening involved a 50 g oral glucose challenge test +/− a 100 g oral glucose tolerance test. Results GDM was diagnosed in 23% pregnancies (n = 104/452). Compared to those without, women with GDM had known risk factors including a higher prepregnancy body mass index (31.1vs 26.3 kg/m2; P &lt; .01) and a prior history of GDM (21.7 vs 5.9%; P &lt; .01). There were no differences in maternal pregnancy complications or neonatal outcomes between groups. Attendance at postpartum glucose testing among women with GDM was poor at 27.9% (29/104). Conclusion These data suggest that women with twin pregnancies share a similar GDM risk profile to those with singleton pregnancies and provide reassuring evidence that current management for GDM twin pregnancies produces similar outcomes to twin pregnancies without GDM.