Search

Background and objective: Data on interaction of patients with artificial intelligence (AI) are limited, primarily derived from small-scale studies, cross-sectional surveys, and qualitative reviews. Most patients have not yet encountered AI in their clinical experience. This study explored patients’ confidence in AI, specifically large language models, after a direct interaction with a chatbot in a clinical setting. Through hands-on experience, the study sought to reduce potential biases due to an anticipated lack of AI experience in a real-world urological patient sample. Methods: A total of 300 patients scheduled for counseling were enrolled from February to July 2024. Participants voluntarily conversed about their medical questions with a GPT-4 powered chatbot, followed by a survey assessing their confidence in clinical capabilities of AI compared with their counseling urologists. Clinical capabilities included history taking, diagnostics, treatment recommendation, anxiety reduction, and time allocation. Key findings and limitations: Of the 292 patients who completed the study, AI was significantly preferred to physicians for consultation time allocation (p

Abstract Medical image classification requires labeled, task-specific datasets which are used to train deep learning networks de novo, or to fine-tune foundation models. However, this process is computationally and technically demanding. In language processing, in-context learning provides an alternative, where models learn from within prompts, bypassing the need for parameter updates. Yet, in-context learning remains underexplored in medical image analysis. Here, we systematically evaluate the model Generative Pretrained Transformer 4 with Vision capabilities (GPT-4V) on cancer image processing with in-context learning on three cancer histopathology tasks of high importance: Classification of tissue subtypes in colorectal cancer, colon polyp subtyping and breast tumor detection in lymph node sections. Our results show that in-context learning is sufficient to match or even outperform specialized neural networks trained for particular tasks, while only requiring a minimal number of samples. In summary, this study demonstrates that large vision language models trained on non-domain specific data can be applied out-of-the box to solve medical image-processing tasks in histopathology. This democratizes access of generalist AI models to medical experts without technical background especially for areas where annotated data is scarce.

The present research investigates the effect of axial relative density on the ultimate load of model piles with different lengths driven in dry sandy soil having small scale model piles.The materials used in this study are divided into three parts they are; sandy soil, steel piles and aluminum pile caps.A total number of 27 model tests are carried out using three relative densities (33%, 60%, and 80%) corresponding to loose, medium and dense sand, respectively. A square section steel solid model piles are used with 18 mm width and (320, 420, 520) mm embedded length (Ld). A 6B distance between the piles center to center is selected to eliminate the effect of group interaction.Two pile groups' configuration (1×2) and (2 × 2) are considered in this study connected by aluminum caps with smooth surface having a thickness of 25 mm.The load applied on the models is measured by a pressure transducer connected to the main line of the hydraulic pressure system and applying up to failure. During all the experimental tests, the loading rate is kept at 3 mm/min. It was founded that relative density has more impact on (2×2) pile group than on (1×2) pile group and single pile. The average rate of increase in the ultimate load from loose to medium is about 15% greater than the average rate of increase from medium to dense. Also, the ultimate load of pile increased about 96% with changing the sand density from loose to medium sand for single and piles group while the average increment is about 81% from medium to dense. The increase in embedded length of pile caused increasing the ultimate load capacity and decreasing the settlement ratio. The average rate of increase in the ultimate load when the embeddedlength changes from 32 cm to 42 cm is about 4% less than the average rate of increase when the embedded length changes from 42 cm to 52 cm. Also, use of (1×2) piles group instead of a single pile, the average rate of increase is about 6% less than the average rate of increase when change from (1×2) group to (2×2) pile group.Finally, when the number of piles changes from single pile to (1×2) piles group, and to (2×2) piles group, the average of increase of ultimate load is about 102.5% and 108.5% respectively.

IntroductionVery rarely, adult NMDAR antibody-associated encephalitis (NMDAR-E) leads to persistent cerebellar atrophy and ataxia. Transient cerebellar ataxia is common in pediatric NMDAR-E. Immune-mediated cerebellar ataxia may be associated with myelin oligodendrocyte glycoprotein (MOG), aquaporin-4 (AQP-4), kelch-like family member 11 (KLHL11), and glutamate kainate receptor subunit 2 (GluK2) antibodies, all of which may co-occur in NMDAR-E. Here, we aimed to investigate the frequency, long-term outcome, and immunological concomitants of ataxia in NMDAR-E.MethodsIn this observational study, patients with definite NMDAR-E with a follow-up of >12 months were recruited from the GENERATE registry. Cases with documented ataxia were analyzed in detail.ResultsIn 12 of 62 patients (19%), ataxia was documented. Bilateral cerebellar ataxia without additional focal CNS findings was found in four (one child and three adults); one of these was previously reported as a case with persistent cerebellar atrophy and ataxia. Two patients with bilateral cerebellar ataxia had additional focal neurological symptoms, optic neuritis and facial palsy. Two patients developed hemiataxia: one with diplopia suggesting brainstem dysfunction and the other probably resulting from cerebellar diaschisis due to contralateral status epilepticus. In all but the one developing cerebellar atrophy, cerebellar ataxia was transient and not associated with a worse long-term outcome. In all five patients with cerebellar ataxia tested, MOG, AQP-4, GluK2, and KLHL11 antibodies were negative. In two additional patients negative for both MOG and AQP-4 antibodies, ataxia was sensory and explained by cervical myelitis as part of multiple sclerosis (MS) manifesting temporal relation to NMDAR-E. One of the patients with bilateral ataxia with focal neurological deficits was also diagnosed with MS upon follow-up. Finally, in two patients, ataxia was explained by cerebral hypoxic damage following circulatory failure during an ICU stay with severe NMDAR-E.DiscussionAtaxia of different types is quite common in NMDAR-E. Cerebellar ataxia in NMDAR-E is mostly transient. NMDAR-E followed by persistent ataxia and cerebellar atrophy is very rare. Cerebellar ataxia in NMDAR-E may not be explained by concomitant KLHL11, MOG, AQP-4, or GluK2 autoimmunity. Of note, ataxia in NMDAR-E may result from treatment complications and, most interestingly, from MS manifesting in temporal association with NMDAR-E.

Summary: Artificial intelligence (AI) methods enable humans to analyse large amounts of data, which would otherwise not be feasibly quantifiable. This is especially true for unstructured visual and textual data, which can contain invaluable insights into disease. The hepatology research landscape is complex and has generated large amounts of data to be mined. Many open questions can potentially be addressed with existing data through AI methods. However, the field of AI is sometimes obscured by hype cycles and imprecise terminologies. This can conceal the fact that numerous hepatology research groups already use AI methods in their scientific studies. In this review article, we aim to assess the contemporaneous use of AI methods in hepatology in Europe. To achieve this, we systematically surveyed all scientific contributions presented at the EASL Congress 2024. Out of 1,857 accepted abstracts (1,712 posters and 145 oral presentations), 6 presentations (∼4%) and 69 posters (∼4%) utilised AI methods. Of these, 55 posters were included in this review, while the others were excluded due to missing posters or incomplete methodologies. Finally, we summarise current academic trends in the use of AI methods and outline future directions, providing guidance for scientific stakeholders in the field of hepatology.

Abstract The WHO guidelines for classifying central nervous system (CNS) tumours are changing considerably with each release. The classification of CNS tumours is uniquely complex among most other solid tumours as it incorporates not just morphology, but also genetic and epigenetic features. Keeping current with these changes across medical fields can be challenging, even for clinical specialists. Large language models (LLMs) have demonstrated their ability to parse and process complex medical text, but their utility in neuro‐oncology has not been systematically tested. We hypothesised that LLMs can effectively diagnose neuro‐oncology cases from free‐text histopathology reports according to the latest WHO guidelines. To test this hypothesis, we evaluated the performance of ChatGPT‐4o, Claude‐3.5‐sonnet, and Llama3 across 30 challenging neuropathology cases, which each presented a complex mix of morphological and genetic information relevant to the diagnosis. Furthermore, we integrated these models with the latest WHO guidelines through Retrieval‐Augmented Generation (RAG) and again assessed their diagnostic accuracy. Our data show that LLMs equipped with RAG, but not without RAG, can accurately diagnose the neuropathological tumour subtype in 90% of the tested cases. This study lays the groundwork for a new generation of computational tools that can assist neuropathologists in their daily reporting practice.

Hypocobalaminemia is common in cats with chronic enteropathy (FCE). However, the disruptions in cobalamin metabolism are not fully understood and may vary across species. Cobalamin is distributed to target tissues via binding to transcobalamin (TC) in blood, which has not been evaluated in cats. Thus, an in-house sandwich-ELISA was established to evaluate serum total TC concentrations in cats with FCE. Surplus sera served to analytically validate the assay, and serum TC concentrations were compared among cats with FCE and other diseases (gastrointestinal neoplasia, cholangiohepatopathy, and other neoplastic or non-neoplastic conditions) and healthy controls. Observed-to-expected ratios for serial dilutions ranged from 72.4 to 145.6% and were 75.1–126.7% for spiking-and-recovery. Intra- and inter-assay variability was p = 0.0067) but not correlated with paired cobalamin concentrations in FCE. Hypertranscobalaminemia predominated with hypercobalaminemia, reaching the highest levels in advanced-stage chronic kidney disease (CKD) cases. TC variations in cobalamin deficiency states with FCE may be linked to inflammation or autoantibodies. This and possible links between serum TC variation in FCE, intracellular cobalamin availability, response to supplementation, and concurrent CKD require further exploration.