xix, 1237 hal; 18 c

In this paper, we introduce InMD-X, a collection of multiple large language models specifically designed to cater to the unique characteristics and demands of Internal Medicine Doctors (IMD). InMD-X represents a groundbreaking development in natural language processing, offering a suite of language models fine-tuned for various aspects of the internal medicine field. These models encompass a wide range of medical sub-specialties, enabling IMDs to perform more efficient and accurate research, diagnosis, and documentation. InMD-X's versatility and adaptability make it a valuable tool for improving the healthcare industry, enhancing communication between healthcare professionals, and advancing medical research. Each model within InMD-X is meticulously tailored to address specific challenges faced by IMDs, ensuring the highest level of precision and comprehensiveness in clinical text analysis and decision support. This paper provides an overview of the design, development, and evaluation of InMD-X, showcasing its potential to revolutionize the way internal medicine practitioners interact with medical data and information. We present results from extensive testing, demonstrating the effectiveness and practical utility of InMD-X in real-world medical scenarios.

In this paper, attention is drawn to the importance of accounting for osmotic pressure when analyzing physiological effects on cellular structures in plasma medicine. Interaction of a weakly ionized plasma jet with a saline solution leads to a detectable changes in the saline's ion-molecular composition and hence changes in the osmotic pressure. This, in turn, leads to a stretching or compression of the membrane, depending on the difference of total external and internal pressures. The selective effect of plasma on cells, observed in experiments, is associated with the change in the mechanical properties of membranes (and thereby, a weakening of their protective properties). Corresponding estimates are given in the article.

Objective To develop an LLM based realtime compound diagnostic medical AI interface and performed a clinical trial comparing this interface and physicians for common internal medicine cases based on the United States Medical License Exam (USMLE) Step 2 Clinical Skill (CS) style exams. Methods A nonrandomized clinical trial was conducted on August 20, 2024. We recruited one general physician, two internal medicine residents (2nd and 3rd year), and five simulated patients. The clinical vignettes were adapted from the USMLE Step 2 CS style exams. We developed 10 representative internal medicine cases based on actual patients and included information available on initial diagnostic evaluation. Primary outcome was the accuracy of the first differential diagnosis. Repeatability was evaluated based on the proportion of agreement. Results The accuracy of the physicians' first differential diagnosis ranged from 50% to 70%, whereas the realtime compound diagnostic medical AI interface achieved an accuracy of 80%. The proportion of agreement for the first differential diagnosis was 0.7. The accuracy of the first and second differential diagnoses ranged from 70% to 90% for physicians, whereas the AI interface achieved an accuracy rate of 100%. The average time for the AI interface (557 sec) was 44.6% shorter than that of the physicians (1006 sec). The AI interface ($0.08) also reduced costs by 98.1% compared to the physicians' average ($4.2). Patient satisfaction scores ranged from 4.2 to 4.3 for care by physicians and were 3.9 for the AI interface Conclusion An LLM based realtime compound diagnostic medical AI interface demonstrated diagnostic accuracy and patient satisfaction comparable to those of a physician, while requiring less time and lower costs. These findings suggest that AI interfaces may have the potential to assist primary care consultations for common internal medicine cases.

This prognostic study assesses the ability of a chatbot to write a history of present illness compared with senior internal medicine residents.

Contained within this volume are the scholarly contributions presented in both oral and poster formats at Fully3D 2023: The 17th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine. This conference convened from July 16-21, 2023, at Stony Brook University in New York. For ease of reference, all papers are organized alphabetically according to the last names of the primary authors. Our heartfelt appreciation goes out to all participants who took the time to submit, present, and revise their work for inclusion in these proceedings. Collectively, we would also like to express our profound gratitude to our generous sponsors, detailed in subsequent pages, who have played an instrumental role in offering awards and facilitating the various conference activities. Additionally, our thanks extend to the diligent reporter who collated invaluable feedback from attendees, which can be found in the pages that follow. September 7, 2023 Fully3D 2023 Co-Chairs: Jerome Liang, Paul Vaska, and Chuan Huang

Albumin is the most abundant circulating protein and provides about 70% of the plasma oncotic power. The molecule also carries many other biological functions (binding, transport and detoxification of endogenous and exogenous compounds, antioxidation, and modulation of inflammatory and immune responses). Hypoalbuminemia is a frequent finding in many diseases, representing usually only a biomarker of poor prognosis rather than a primary pathophysiological event. Despite that, albumin is prescribed in many conditions based on the assumption that correction of hypoalbuminemia would lead to clinical benefits for the patients. Unfortunately, many of these indications are not supported by scientific evidence (or have been even disproved), so that a large part of albumin use is nowadays still inappropriate. Decompensated cirrhosis is the clinical area where albumin administration has been extensively studied and solid recommendations can be made. Besides prevention and treatment of acute complications, long-term albumin administration in patients with ascites has emerged in the last decade has a potential new disease-modifying treatment. In non-hepatological settings, albumin is widely used for fluid resuscitation in sepsis and critical illnesses, with no clear superiority over crystalloids. In many other conditions, scientific evidence supporting albumin prescription is weak or even absent. Thus, given its high cost and limited availability, action is needed to avoid the use of albumin for inappropriate and futile indications to ensure its availability in those conditions for which albumin has been demonstrated to have a real effectiveness and an advantage for the patient.

Monte Carlo (MC) techniques are currently deemed the gold standard for internal dosimetry, since the simulations can perform full radiation transport and reach a precision level not attainable by analytical methods. In this study, a custom voxelized particle source was developed for the TOPAS-MC toolkit to be used for internal dosimetry purposes. The source was designed to allow the use of clinical functional scans data to simulate events that reproduce the patient-specific tracer biodistribution. Simulation results are very promising, showing that this can be a first step towards the extension of TOPAS-MC to nuclear medicine applications. In the future more studies are needed to further ascertain the applicability and accuracy of the developed routines.

Background Second victims, defined as healthcare team members being traumatised by an unanticipated clinical event or outcome, are frequent in healthcare. Evidence of this phenomenon in Germany, however, is sparse. Recently, we reported the first construction and validation of a German questionnaire. This study aimed to understand this phenomenon better in a sample of young (<= 35 years) German physicians. Methods The electronic questionnaire (SeViD-I survey) was administered for 6 weeks to a sample of young physicians in training for internal medicine or a subspecialty. All physicians were members of the German Society of Internal Medicine. The questionnaire had three domains - general experience, symptoms, and support strategies - comprising 46 items. Binary logistic regression models were applied to study the influence of various independent factors on the risk of becoming a second victim, the magnitude of symptoms and the time to self-perceived recovery. Results The response rate was 18% (555/3047). 65% of the participants were female, the mean age was 32 years. 59% experienced second victim incidents in their career so far and 35% during the past 12 months. Events with patient harm and unexpected patient deaths or suicides were the most frequent key incidents. 12% of the participants reported that their self-perceived time to full recovery was more than 1 year or have never recovered. Being female was a risk factor for being a second victim (odds ratio (OR) 2.5) and experiencing a high symptom load (OR 2). Working in acute care was promoting a shorter duration to self-perceived recovery (OR 0.5). Support measures with an exceptionally high approval among second victims were the possibility to discuss emotional and ethical issues, prompt debriefing/crisis intervention after the incident and a safe opportunity to contribute insights to prevent similar events in the future. Conclusion The second victim phenomenon is frequent among young German physicians in internal medicine. In general, these traumatic events have a potentially high impact on physician health and the care they deliver. A better understanding of second victim traumatisations in Germany and broad implementation of effective support programs are warranted.

These proceedings contain the presented papers at the 16th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine 2021, which was to be held in Leuven, Belgium, but had to be transformed in a virtual meeting because of the covid-19 pandemic. Fully3D is one of the major forums for presentation of new results in the field of 3D image reconstruction, primarily with applications in x-ray computer tomography, PET and SPECT. Over the life-time of the meeting the focus has shifted to reflect recent developments in the field. Many of the major developments in fully 3D PET and SPECT imaging were first presented at Fully3D, as were the key results for analytic reconstruction methods in cone beam x-ray CT. Similarly, in the 2021 meeting a broad range of topics has been presented. As expected, deep learning methods are being used increasingly: they are used as post-processing tools or incorporated in the reconstruction, often aiming at improved noise and/or artifact suppression.

Background New virtual resources (“novel resources”) have been incorporated into medical education. No recent large studies about their use and perception among internal medicine (IM) residents exist. Objective Characterize the use and perceived helpfulness of educational resources. Design Nationwide survey from December 2019 to March 2020. Participants IM residents in the USA. Main measures Residents were surveyed on their use and their perceived helpfulness of resources for both attaining general medical knowledge and for point-of-care (POC) learning. Traditional resources included board review resources, clinical experience, digital clinical resources (e.g., UpToDate), journal articles, pocket references, professional guidelines, textbooks, and residency curricula. Novel resources included Twitter, video streaming platforms (e.g., YouTube), online blogs, podcasts, and Wikipedia. Key results We had 662 respondents from 55 residency programs across 26 states. On average, residents used 9 total resources (7 traditional and 2 novel). Digital clinical resources and clinical experience were used by all residents and found helpful by the highest percentage of residents (96% and 94%, respectively). Journal articles were next (used by 90%), followed by board review resources and residency curricula (both used by 85%). Their perceived helpfulness varied, from 90% for board review resources, to 66% for journal articles and 64% for residency curricula, the lowest perceived helpfulness of any traditional resource. Podcasts and video streaming platforms were used as frequently as textbooks (58–59%), but were rated as helpful more frequently (75% and 82% vs 66%, respectively). Conclusions Digital clinical resources, video streaming platforms, and podcasts were perceived as helpful, underscoring the importance of ensuring their integration into medical education to complement clinical experience and other traditional resources which remain highly valued by residents. Importance Our findings can inform residency programs as they transition to virtual curricula in the wake of the COVID-19 pandemic.

We provide additional statistical background for the methodology developed in the clinical analysis of knee osteoarthritis in "A Precision Medicine Approach to Develop and Internally Validate Optimal Exercise and Weight Loss Treatments for Overweight and Obese Adults with Knee Osteoarthritis" (Jiang et al. 2020). Jiang et al. 2020 proposed a pipeline to learn optimal treatment rules with precision medicine models and compared them with zero-order models with a Z-test. The model performance was based on value functions, a scalar that predicts the future reward of each decision rule. The jackknife (i.e., leave-one-out cross validation) method was applied to estimate the value function and its variance of several outcomes in IDEA. IDEA is a randomized clinical trial studying three interventions (exercise (E), dietary weight loss (D), and D+E) on overweight and obese participants with knee osteoarthritis. In this report, we expand the discussion and justification with additional statistical background. We elaborate more on the background of precision medicine, the derivation of the jackknife estimator of value function and its estimated variance, the consistency property of jackknife estimator, as well as additional simulation results that reflect more of the performance of jackknife estimators. We recommend reading Jiang et al. 2020 for clinical application and interpretation of the optimal ITR of knee osteoarthritis as well as the overall understanding of the pipeline and recommend using this article to understand the underlying statistical derivation and methodology.

The aim in this systematic review was to determine the effect of point-of-care ultrasound (POCUS) on the clinical decision-making process and patient outcomes in adults admitted to the general medicine ward. A comprehensive search was performed in MEDLINE (Ovid), EMBASE (Ovid), PubMed, the Cochrane Library, ClinicalTrials.gov, Scopus, LILACS and Cinahl. Articles had to fulfill the inclusion criteria of randomised or non-randomised studies assessing the impact of POCUS on the diagnosis, management, length of hospital stay or mortality of patients admitted to the internal medicine ward. Six studies were included involving a total of 1836 patients. The influence of POCUS on the diagnosis was reported as a change in the main diagnosis or the addition of a relevant diagnosis in up to 18% and 24% of the cases, respectively. Impact on the management plan was reported in 37% to 52.1% of the participants. Three studies documented the impact of POCUS on the length of stay. Two of them reported no difference between groups, and the other reported a significant reduction of 1 d of the hospital stay. In conclusion, POCUS appears to have positive effects on the clinical decision-making process with impacts on optimal patient management and possible reduction in the hospital length of stay.

Purpose: Comprehensive legal medicine documentation includes both an internal but also an external examination of the corpse. Typically, this documentation is conducted manually during conventional autopsy. A systematic digital documentation would be desirable, especially for the external examination of wounds, which is becoming more relevant for legal medicine analysis. For this purpose, RGB surface scanning has been introduced. While a manual full surface scan using a handheld camera is timeconsuming and operator dependent, floor or ceiling mounted robotic systems require substantial space and a dedicated room. Hence, we consider whether a mobile robotic system can be used for external documentation. Methods: We develop a mobile robotic system that enables full-body RGB-D surface scanning. Our work includes a detailed configuration space analysis to identify the environmental parameters that need to be considered to successfully perform a surface scan. We validate our findings through an experimental study in the lab and demonstrate the system's application in a legal medicine environment. Results: Our configuration space analysis shows that a good trade-off between coverage and time is reached with three robot base positions, leading to a coverage of 94.96 %. Experiments validate the effectiveness of the system in accurately capturing body surface geometry with an average surface coverage of 96.90 +- 3.16 % and 92.45 +- 1.43 % for a body phantom and actual corpses, respectively. Conclusion: This work demonstrates the potential of a mobile robotic system to automate RGB-D surface scanning in legal medicine, complementing the use of post-mortem CT scans for inner documentation. Our results indicate that the proposed system can contribute to more efficient and autonomous legal medicine documentation, reducing the need for manual intervention.

Model Medicine is the science of understanding, diagnosing, treating, and preventing disorders in AI models, grounded in the principle that AI models -- like biological organisms -- have internal structures, dynamic processes, heritable traits, observable symptoms, classifiable conditions, and treatable states. This paper introduces Model Medicine as a research program, bridging the gap between current AI interpretability research (anatomical observation) and the systematic clinical practice that complex AI systems increasingly require. We present five contributions: (1) a discipline taxonomy organizing 15 subdisciplines across four divisions -- Basic Model Sciences, Clinical Model Sciences, Model Public Health, and Model Architectural Medicine; (2) the Four Shell Model (v3.3), a behavioral genetics framework empirically grounded in 720 agents and 24,923 decisions from the Agora-12 program, explaining how model behavior emerges from Core--Shell interaction; (3) Neural MRI (Model Resonance Imaging), a working open-source diagnostic tool mapping five medical neuroimaging modalities to AI interpretability techniques, validated through four clinical cases demonstrating imaging, comparison, localization, and predictive capability; (4) a five-layer diagnostic framework for comprehensive model assessment; and (5) clinical model sciences including the Model Temperament Index for behavioral profiling, Model Semiology for symptom description, and M-CARE for standardized case reporting. We additionally propose the Layered Core Hypothesis -- a biologically-inspired three-layer parameter architecture -- and a therapeutic framework connecting diagnosis to treatment.

The manual outlining of hepatic metastasis in (US) ultrasound acquisitions from patients suffering from pancreatic cancer is common practice. However, such pure manual measurements are often very time consuming, and the results repeatedly differ between the raters. In this contribution, we study the in-depth assessment of an interactive graph-based approach for the segmentation for pancreatic metastasis in US images of the liver with two specialists in Internal Medicine. Thereby, evaluating the approach with over one hundred different acquisitions of metastases. The two physicians or the algorithm had never assessed the acquisitions before the evaluation. In summary, the physicians first performed a pure manual outlining followed by an algorithmic segmentation over one month later. As a result, the experts satisfied in up to ninety percent of algorithmic segmentation results. Furthermore, the algorithmic segmentation was much faster than manual outlining and achieved a median Dice Similarity Coefficient (DSC) of over eighty percent. Ultimately, the algorithm enables a fast and accurate segmentation of liver metastasis in clinical US images, which can support the manual outlining in daily practice.