Hasil untuk "Diseases of the digestive system. Gastroenterology"

Kei Ito, Akira Endo, Hiromasa Hoshi et al.

ABSTRACT Background Although surgical treatment is the primary measure for patients with perforated peptic ulcer (PPU), nonoperative management (NOM) has become a common alternative. However, risk score models predicting the success of NOM based on the analysis of a large number of patients remain scarce. We developed a clinically applicable scoring system to predict the success of NOM in patients with PPU using data from a large cohort. Method We analyzed data of the Perforated Peptic ulcer Analyzing Project (PPAP), which was a retrospective survey of adult patients with PPU between January 2011 to December 2022. The successful NOM case was defined as patients who survived until hospital discharge without requiring surgery. Factors associated with NOM were identified using a multivariable logistic regression analysis, and a scoring system to predict NOM was developed by weighting these factors based on the regression coefficients. Result Of 702 potentially eligible patients, 584 were treated with NOM, of which 130 patients (22.2%) were treated successfully. Age, sex, body temperature, heart rate, the extent of peritoneal irritation signs, C reactive protein, spread of ascites, and sepsis were included in the final model. Using these variables, we developed the scoring system named PPAP score, which had favorable discriminating ability with the area under receiving operating characteristic curve of 0.799. When the cut‐off was set to 56, the sensitivity and the specificity were 0.738 and 0.722, respectively. Conclusion A predictive scoring model was developed. However, external validation of the model is required to confirm its clinical applicability.

Javier Crespo, Ana Enériz, Paula Iruzubieta et al.

Background: Artificial intelligence (AI) has emerged as a disruptive innovation in medicine, yet its adoption within gastroenterology remains limited and poorly characterized. We aimed to examine knowledge, practical applications, perceived barriers, and expectations regarding AI among gastroenterology specialists in Spain. Methods: We conducted a cross-sectional observational study using a structured online survey distributed by the Spanish Society of Digestive Pathology (SEPD) in 2025. The questionnaire collected sociodemographic data, patterns of AI use, perceptions, and educational needs. Descriptive statistics and multivariable models were applied. Results: Among 283 respondents (mean age 44.6 +/- 9.7 years), 87.5% acknowledged AI as a transformative tool, but only 60.2% (95% CI: 54.3-66.1%) reported using it, mostly outside institutional frameworks. Notably, 80.2% of users initiated AI use within the past year. Independent predictors of frequent use included previous training (OR=2.44), employment in university hospitals (OR=2.14), and younger age (OR=1.36 per 5-year decrease). Main barriers were lack of training (61%), absence of institutional strategies (46%), and ethical concerns (50%). While 93.8% agreed that AI training programmes are necessary, only 18.4% had received formal training. Conclusions: A substantial gap exists between the favorable perception of AI and its actual integration into clinical practice within Spanish gastroenterology. The rapid adoption outside institutional frameworks underscores the urgent need for accredited training programmes and governance standards led by scientific societies.

J. Fonseca

This Special Issue in Nutrients, "Dietary and Nutritional Therapies to Improve Digestive Disorders," focuses on the fundamental and increasingly recognized relationship between nutrition and digestive diseases [...].

Ku Bean Jeong, Hee Seok Moon, Kyung Ryun In et al.

Abstract Background The incidence of lower gastrointestinal bleeding is on the rise, prompting the creation of various scoring systems to forecast patient’s outcomes. But there is no single unified scoring system and these scoring systems clinical data are small and not worldwide. Aims To evaluate how different scoring systems predict mortality and prolonged hospital stay (≥ 10 days). Methods A retrospective review was conducted on the medical records of 4417 patients who presented with hematochezia at the emergency department from January 2016 to December 2022. We evaluated the predictive accuracy of various scoring systems for 30-day mortality and prolonged hospital stay (≥ 10 days) by analyzing the areas under the receiver-operating characteristic curves, taking into account factors such as patient age, laboratory findings, and comorbidities (ABC); AIMS 65; Glasgow-Blatchford; Oakland; Rockall(pre-endoscopy); SHA2PE; and CHAMPS scores. Results We analyzed data from 1000 patients (mean age 66 years, 56.1% men, mean hospital stay 9.4 days) with lower gastrointestinal bleeding confirmed by any other means including DRE, colonoscopy and CT. The 30-day mortality rate was 3.7%. The primary etiologies of lower gastrointestinal bleeding were identified as ischemic colitis and diverticular bleeding, accounting for 18.8% and 18.5% of cases, respectively. In terms of forecasting 30-day mortality, the AIMS 65, CHAMPS, and ABC scoring systems demonstrated superior performance (p < 0.001). For predicting prolonged hospital stay, the SHA2PE score exhibited the highest accuracy among all evaluated systems (p < 0.001). Conclusions The newly developed scoring systems demonstrated superior accuracy in forecasting outcomes for patients with lower gastrointestinal bleeding, and the results of this study demonstrate that these scoring systems can be applied in clinical practice.

Nariman Naderi, Seyed Amir Ahmad Safavi-Naini, Thomas Savage et al.

This study evaluated self-reported response certainty across several large language models (GPT, Claude, Llama, Phi, Mistral, Gemini, Gemma, and Qwen) using 300 gastroenterology board-style questions. The highest-performing models (GPT-o1 preview, GPT-4o, and Claude-3.5-Sonnet) achieved Brier scores of 0.15-0.2 and AUROC of 0.6. Although newer models demonstrated improved performance, all exhibited a consistent tendency towards overconfidence. Uncertainty estimation presents a significant challenge to the safe use of LLMs in healthcare. Keywords: Large Language Models; Confidence Elicitation; Artificial Intelligence; Gastroenterology; Uncertainty Quantification

N.N. Sharafutdinova, R. Nadyrgulov, M. Borisova et al.

Significance. Diseases of the digestive system rank 5th to 6th in the structure of disease prevalence in the Russian Federation. They are also the third most frequent reason for hospitalization in multidisciplinary hospitals and one of the five leading causes of mortality, particularly among working-age population. This defines their significant clinical, social, and economic impact. The analysis of clinical and statistical characteristics of patients with digestive diseases is essential for planning organizational and preventive measures based on their prevalence within specific demographic groups. Purpose: To analyze clinical and statistical data on patients with diseases of the digestive system who received inpatient medical care for subsequent use of the analysis results in planning specialized medical care. Material and methods. A study on clinical and statistical characteristics of patients who received inpatient treatment at the gastroenterology and gastrointestinal surgery departments of a multidisciplinary hospital in Ufa was conducted. The analysis was based on data of the Republican Medical Information and Analytical System of the Republic of Bashkortostan (RMIS RB) and Federal Statistical Form No. 14 for the year 2023. A comparative analysis of patients of the two departments by sex, age, disease, and mode of admission was performed and the rate of hospitalization by month and day of the week was evaluated. The study employed statistical and analytical research methods. Statistical processing was performed with the help of conventional statistical methods using the SPSS Statistics and Microsoft Excel software packages. Differences were considered statistically significant at a probability level of 95% (p<0.05) and 99% (p<0.01) Results. Among patients of the gastroenterology department, males accounted for 46.5%, and 49.7% in the gastrointestinal surgery department. Although the 60-70-year group was predominant among the hospitalized patients, the proportion of patients aged 50-59 and 40-49 remained high. The leading disease among patients hospitalized in the gastroenterology department was pancreatitis (52.0%), while for those hospitalized in the gastrointestinal surgery department, it was cholelithiasis (41.9%). The proportion of emergency admissions to the gastroenterology department equaled to 32.4%, with one third of these patients arriving by ambulance. In the gastrointestinal surgery department, the number of patients admitted on an emergency basis was twice as high, accounting for 66.7% of patients. The highest number of hospitalizations due to diseases of the digestive system occurred during the spring and autumn months. Discussion. The study found that the proportion of male patients was lower than that of females in both departments. Furthermore, 60.3% of the hospitalized in the gastroenterology department and 57.5% in the gastrointestinal surgery department were patients of working age (under 60 years). Among the age groups, patients aged 60 to 70 warrant particular attention, accounting for approximately one-quarter of all hospitalizations: 24.3% in the gastroenterology and 25.0% in the gastrointestinal surgery departments. This is associated with the chronic progression of many digestive diseases, especially in older age groups. A high proportion of patients aged 40-49 and 50-59 substantiates the need for enhanced preventive measures targeting these cohorts, which can reduce the frequency of chronic disease exacerbations. The results indicate that every third patient is admitted to the gastroenterology department on an emergency basis, while in the gastrointestinal surgery department, this figure exceeds 60.0%. This suggests insufficient effectiveness of outpatient management, non-adherence to medical recommendations, and failure to attend scheduled follow-up appointments. Conclusion. The study determined that pancreatitis, liver cirrhosis, ulcerative and other forms of colitis, and hernias were the most common conditions among patients with diseases of the digestive system who received inpatient care. One-third of patients in the gastroenterology department and two-thirds in the gastrointestinal surgery department were admitted on an emergency basis. The characterization of patients with digestive diseases by mode of admission, combined with data on the dynamics in hospitalizations by month and day of the week, provides an informational basis for developing measures to improve the organization of medical care delivery to this patient group Keywords: diseases of the digestive system; inpatient care; clinical and statistical characteristics; patients.

G. Camci, Zeliha Cengiz

Purpose: This study aimed to examine the use of herbal products and attitudes toward healthy eating among individuals with digestive system diseases. Method: A cross-sectional study was conducted with 215 patients hospitalized in internal medicine-gastroenterology wards. Data on demographics, medical history, herbal product use, and attitudes toward healthy eating were collected using a questionnaire and the Attitude Scale Toward Healthy Eating. Findings: Among participants, % 36.3 reported bloating as the most common digestive issue. Herbal products were used by % 65.1, with mineral water (% 55.4) and Turkish coffee (% 27.7) being the most common. The mean healthy eating attitude score was 66.07 ± 8.38, indicating moderate positive attitudes. Scores varied based on factors such as digestive problems, meal regularity, eating speed, and frequency of herbal product use, but not on herbal product use itself. Conclusion and Recommendations: While herbal product use was common, it did not directly influence attitudes toward healthy eating. Dietary habits and digestive health had a greater impact on these attitudes. Future studies should delve deeper into the relationship between herbal product use and dietary habits, and further research is recommended on the efficacy and safety of herbal products for improving digestive health.

I. N. Banin, S. Cherkasov

The role of patient, especially at early stages of disease, is decisive one in relation to disease outcome. Therefore, it is extremely important to analyze typical and deviant forms of one's behavior in interaction with health care system. The purpose of the study is to analyze models of behavior of patients with diseases of digestive system in interaction with the health care system. Three comparison groups were formed for analysis. The first one included patients undergoing planned out-patient treatment (n=531), the second one patients undergoing planned in-patient treatment (specialized Gastroenterology Department) (n=822). The third one patients admitted for in-patient treatment at the Emergency Medical Care Hospital on emergency basis (n=147) on account of complications of primary disease (pathology of digestive organs). The groups were compared using such non-parametric criteria as the association coefficient (Cass) and the 2 criterion. It is established that at choosing physician, the most frequently used source of information by patients is opinions of ones who have already was treated by particular physician. In all three groups, this source of information was used most frequently. The second source most frequently used was one's own previous experience of visiting this physician. The third most important criterion of choice was patient own opinion. At that, no reliable differences in using of information sources by patients included in first and second groups were found. The significant differences in behavior models were found among patients of third group that significantly affects possibility of complications requiring emergency hospitalization. The main criterion for choosing medical organization, if such an opportunity exists, is availability of competent specialists that is considered much more important than good reputation of medical organization. The geographical accessibility is also of great importance for patients, especially for ones receiving medical care in out-patient conditions. The specific characteristics of the models of behavior patterns of patients in each group should to determine other general characteristics of behavior associated with interaction of patient and the health care system.

G. Kozhonazarova, O. Z. Uzakov, E. S. Alymbaev et al.

The state of health of children represents serious medical social problem. The data of official statistics and special studies everywhere register increasing of general morbidity and disability, including area of digestive system, chronic forms of diseases in structure of children pathology, share of congenital development defects and genetic deviations. The study was carried out to investigate and analyze structure of morbidity of digestive system in children based on data of appealability for hospitalization in the gastroenterology and hepatology Department of the National Center of Maternity and Childhood Care (Kyrgyzstan). The data from reports of mentioned gastroenterology and hepatology Department in 2019-2024 were analyzed. It was established that across all nosological units that are subjected to hospitalization in hospital, occurs steady increasing that indicates to severity of disease. The facts of increase of morbidity of gastrointestinal tract and hepatobiliary system require further investigation and scientific research to find causal factor.

Yukinori Mashima, Masatoshi Tanigawa, Hideto Yokoi

This study focused on the heterogeneity in progress notes written by physicians or nurses. A total of 806 days of progress notes written by physicians or nurses from 83 randomly selected patients hospitalized in the Gastroenterology Department at Kagawa University Hospital from January to December 2021 were analyzed. We extracted symptoms as the International Classification of Diseases (ICD) Chapter 18 (R00–R99, hereinafter R codes) from each progress note using MedNER-J natural language processing software and counted the days one or more symptoms were extracted to calculate the extraction rate. The R-code extraction rate was significantly higher from progress notes by nurses than by physicians (physicians 68.5% vs. nurses 75.2%; p = 0.00112), regardless of specialty. By contrast, the R-code subcategory R10–R19 for digestive system symptoms (44.2 vs. 37.5%, respectively; p = 0.00299) and many chapters of ICD codes for disease names, as represented by Chapter 11 K00–K93 (68.4 vs. 30.9%, respectively; p < 0.001), were frequently extracted from the progress notes by physicians, reflecting their specialty. We believe that understanding the information heterogeneity of medical documents, which can be the basis of medical artificial intelligence, is crucial, and this study is a pioneering step in that direction.

Giovanni Cammarota, G. Ianiro

N. Gubergrits, L. Zhuravlyova, N. Byelyayeva

Professor V. M. Khvorostinka, a Doctor of Medical Sciences, was an Honored Worker of Science and Technology of Ukraine (1990). At Kharkiv National Medical University, Volodymyr Mykolayovych Khvorostinka worked his way up from clinical resident (1968—1969), assistant (1969—1978), associate professor (1978—1982) to the Head of the Department of Faculty Therapy (1982—2009). He was a talented student and a follower of prominent scientists who headed the Department of Faculty Therapy throughout years. V. M. Khvorostinka effectively combined scientific activity with pedagogical, medical, and social work. He contributed significantly to the training and education of numerous generations of highly qualified physicians in practical health care and science. On his initiative, specialized gastroenterology, rheumatology, cardiology, and endocrinology departments have been established in Kharkiv Regional Clinical Hospital, where he worked as curator and consultant. Approbation of both domestic and foreign medications had been routinely carried out on this clinical base. V. M. Khvorostinka was a member of the European Regional Association of Gastroenterologists and the European Association of Diabetologists. The majority of the scientific works at the Department of Faculty Therapy under the leadership of V. M. Khvorostinka were devoted to the new methods of diagnosis and treatment of chronic liver diseases and the role of the liver in the development of atherosclerosis, particularly in diabetes. Thorough research was conducted on the new methods of early diagnosis, variants of the course, and methods of pathogenetic therapy for chronic liver diseases of alcoholic origin. Professor V. M. Khvorostinka published more than 560 research papers and several manuals and monographs. Under the supervision of V. M. Khvorostinka, two doctoral dissertations and 20 candidate dissertations were defended. V. M. Khvorostinka was not only an outstanding scientist, teacher, and highly qualified doctor but also a talented and creative person. He painted amazing pictures in various genres. The paintings of V. M. Khvorostinka are still exhibited in the art gallery at Kharkiv National Medical University.

N. Gubergrits, N. Byelyayeva, K. Linevska

Nina Fedorivna Deyneko is a Doctor of Medical Sciences, Professor, an honored worker of science and technology of Ukraine, an outstanding health care worker, a labor veteran, and an academician of the International Academy of Sciences in apiphytotherapy and beekeeping. Under Prof. N. F. Deyneko’s supervision, 86 clinical residents, including 34 foreign ones, and postgraduate students were taught during her tenure at the Kharkiv departments of therapy and gastroenterology. Additionally, 25 candidate dissertations and one doctorate thesis were defended. She served as the head of the Kharkiv Medical Institute’s state examination commission for 15 years, founded and presided over the Kharkiv Medical Society of Gastroenterologists for 20 years, was a member of the All‑Union Society of Gastroenterologists’ Presidium of the Board for more than 20 years, and worked as the deputy chairman of the board of the Ukrainian Republican Society of Gastroenterologists. More than 200 research publications, four monographs, textbooks, and copyright certificates have been produced by Nina Fedorivna. Kharkiv Railway Clinical Hospital №1 became the clinical base of the Department of Gastroenterology of the Ukrainian Institute for Advanced Medical Studies, which was headed by Professor Nina Fedorivna Deyneko. Contemporary diagnostic equipment made it possible to perform diagnosis and treatment of patients with pathologies of the gastrointestinal, liver, biliary and pancreatic pathologies. The goal of methodological and pedagogical work was to optimize the educational process by using modern techniques for tracking students’ practical skills and initial, intermediate, and final levels of learning. Four‑ and three‑ month specialty cycles, as well as pre‑certification and theme improvement in the field of gastroenterology, were implemented during the first five years of the department’s existence. In addition to receiving 20 copyright certificates and publishing over 300 research papers, department staff members have also developed 62 rationalization recommendations that have been put into practice in the healthcare industry. The staff of the department participated in many scientific forums, including those conducted abroad.

Ben-Gang Zhou, Ming-Wen Guo, Li-Juan Zhang et al.

Background: The efficacy of the 14-day esomeprazole–amoxicillin (EA) dual therapy in eradicating Helicobacter pylori ( H. pylori ) has been widely discussed previously. Vonoprazan, a novel potassium-competitive acid blocker, presents rapid, potent, and long-lasting acid inhibitory effects compared to esomeprazole. However, there is currently a scarcity of direct comparisons between the 10-day vonoprazan–amoxicillin (VA) and the 14-day EA dual therapy for H. pylori eradication. Objectives: This study aimed to compare the efficacy and safety of the 10-day VA and the 14-day EA dual therapy for H. pylori first-line eradication. Design: This study was a prospective, multicenter, open-label, randomized controlled trial. Methods: The study was conducted at 10 hospitals in China. In total, 570 newly diagnosed H. pylori -infected patients were recruited from April 2023 to February 2024. These patients were randomly assigned to either the 10-day VA group (vonoprazan 20 mg twice daily + amoxicillin 1000 mg three times daily) or the 14-day EA group (esomeprazole 20 mg four times daily + amoxicillin 750 mg four times daily). The primary outcome was the eradication rate, with secondary outcomes including adverse events and compliance. Results: The 10-day VA regimen outperformed the 14-day EA regimen in terms of eradication rates in intention-to-treat (ITT) analysis (85.4% vs 76.7%, p  = 0.008), modified ITT analysis (90.7% vs 84.8%, p  = 0.036), and per-protocol (PP) analysis (91.1% versus 85.5%, p  = 0.047). The non-inferiority p -values in all three analyses were less than 0.001. No statistically significant difference was observed in the incidence of adverse events between the two groups (9.1% vs 11.7%, p  = 0.308). The 10-day VA regimen demonstrated higher compliance compared to the 14-day EA regimen ( p  = 0.006). Conclusion: The 10-day VA dual therapy showed a satisfactory eradication rate of 91.1% (PP analysis), demonstrating good safety and better compliance compared to the 14-day EA dual therapy as the first-line eradication. Trial registration: This trial was registered in the Chinese Clinical Trial Registry (registration number: ChiCTR2300070475) on April 12, 2023.

Pimsiri Sripongpun, Apichat Kaewdech, Prowpanga Udompap et al.

Background &amp; Aims: The new nomenclature of steatotic liver disease (SLD) was recently launched with sub-classifications of metabolic dysfunction-associated SLD (MASLD), MASLD with increased alcohol intake (MetALD), and alcohol-related liver disease (ALD). Herein, we aimed to evaluate the characteristics and long-term outcomes associated with these subgroups and the utility of non-invasive biomarkers. Methods: Using NHANES III (the third National Health and Nutrition Examination Survey) and linked mortality data, all adult participants with available ultrasonographic liver steatosis status were included. Those with viral hepatitis, incomplete data on alcohol consumption, cardiometabolic risk, and missing data that hindered Steatosis-associated Fibrosis Estimator (SAFE) score calculation were excluded. The characteristics of those without SLD (no steatosis on ultrasound), MASLD, MetALD, and ALD were compared. Overall survival (OS) was determined and SAFE score strata were applied to SLD subgroups. Results: A total of 9,939 participants were eligible; 64% had no SLD, while 30%, 2.3%, and 1% had MASLD, MetALD, and ALD, respectively. A higher proportion of men, as well as active smokers, was observed in the MetALD and ALD groups compared to the MASLD group. Diabetes was more prevalent in the MASLD group than in the MetALD and ALD groups. The ALD subgroup had significantly lower OS than the MASLD group (p = 0.004), but the MetALD did not (p = 0.165). SAFE score strata meaningfully differentiated OS of all SLD subgroups. Conclusions: MASLD accounted for the largest proportion of SLD. MetALD shared the characteristics of both MASLD and ALD. The ALD subgroup had a significantly lower OS than the MASLD subgroup but there was no difference between MetALD and MASLD. The SAFE score can be used to stratify long-term outcomes in all SLD subgroups. Impact and implications:: “Steatotic liver disease (SLD)” is a recently introduced term covering three subgroups: MASLD (metabolic dysfunction-associated SLD), MetALD (MASLD with increased alcohol intake), and ALD (alcohol-related liver disease). We explored the characteristics and outcomes of these subgroups among the US population. We found that MASLD was far more common than MetALD and ALD, but all subgroups shared cardiometabolic risk factors. The ALD subgroup has the worst survival, pointing to the synergistic effect of alcohol and metabolic dysfunction. In addition, the SAFE (Steatosis-associated Fibrosis Estimator) score might be a useful non-invasive test to stratify long-term risk in all three SLD subgroups.

Bing Chen, MD, David L. Diehl, MD, FASGE

Moez Dawood, Ben Heavner, Marsha M. Wheeler et al.

Rare diseases are collectively common, affecting approximately one in twenty individuals worldwide. In recent years, rapid progress has been made in rare disease diagnostics due to advances in DNA sequencing, development of new computational and experimental approaches to prioritize genes and genetic variants, and increased global exchange of clinical and genetic data. However, more than half of individuals suspected to have a rare disease lack a genetic diagnosis. The Genomics Research to Elucidate the Genetics of Rare Diseases (GREGoR) Consortium was initiated to study thousands of challenging rare disease cases and families and apply, standardize, and evaluate emerging genomics technologies and analytics to accelerate their adoption in clinical practice. Further, all data generated, currently representing ~7500 individuals from ~3000 families, is rapidly made available to researchers worldwide via the Genomic Data Science Analysis, Visualization, and Informatics Lab-space (AnVIL) to catalyze global efforts to develop approaches for genetic diagnoses in rare diseases (https://gregorconsortium.org/data). The majority of these families have undergone prior clinical genetic testing but remained unsolved, with most being exome-negative. Here, we describe the collaborative research framework, datasets, and discoveries comprising GREGoR that will provide foundational resources and substrates for the future of rare disease genomics.

Seyed Amir Ahmad Safavi-Naini, Shuhaib Ali, Omer Shahab et al.

Background and Aims: This study evaluates the medical reasoning performance of large language models (LLMs) and vision language models (VLMs) in gastroenterology. Methods: We used 300 gastroenterology board exam-style multiple-choice questions, 138 of which contain images to systematically assess the impact of model configurations and parameters and prompt engineering strategies utilizing GPT-3.5. Next, we assessed the performance of proprietary and open-source LLMs (versions), including GPT (3.5, 4, 4o, 4omini), Claude (3, 3.5), Gemini (1.0), Mistral, Llama (2, 3, 3.1), Mixtral, and Phi (3), across different interfaces (web and API), computing environments (cloud and local), and model precisions (with and without quantization). Finally, we assessed accuracy using a semiautomated pipeline. Results: Among the proprietary models, GPT-4o (73.7%) and Claude3.5-Sonnet (74.0%) achieved the highest accuracy, outperforming the top open-source models: Llama3.1-405b (64%), Llama3.1-70b (58.3%), and Mixtral-8x7b (54.3%). Among the quantized open-source models, the 6-bit quantized Phi3-14b (48.7%) performed best. The scores of the quantized models were comparable to those of the full-precision models Llama2-7b, Llama2--13b, and Gemma2-9b. Notably, VLM performance on image-containing questions did not improve when the images were provided and worsened when LLM-generated captions were provided. In contrast, a 10% increase in accuracy was observed when images were accompanied by human-crafted image descriptions. Conclusion: In conclusion, while LLMs exhibit robust zero-shot performance in medical reasoning, the integration of visual data remains a challenge for VLMs. Effective deployment involves carefully determining optimal model configurations, encouraging users to consider either the high performance of proprietary models or the flexible adaptability of open-source models.

Snehasis Banerjee, Sayan Paul, Ruddradev Roychoudhury et al.

This article presents Teledrive, a telepresence robotic system with embodied AI features that empowers an operator to navigate the telerobot in any unknown remote place with minimal human intervention. We conceive Teledrive in the context of democratizing remote care-giving for elderly citizens as well as for isolated patients, affected by contagious diseases. In particular, this paper focuses on the problem of navigating to a rough target area (like bedroom or kitchen) rather than pre-specified point destinations. This ushers in a unique AreaGoal based navigation feature, which has not been explored in depth in the contemporary solutions. Further, we describe an edge computing-based software system built on a WebRTC-based communication framework to realize the aforementioned scheme through an easy-to-use speech-based human-robot interaction. Moreover, to enhance the ease of operation for the remote caregiver, we incorporate a person following feature, whereby a robot follows a person on the move in its premises as directed by the operator. Moreover, the system presented is loosely coupled with specific robot hardware, unlike the existing solutions. We have evaluated the efficacy of the proposed system through baseline experiments, user study, and real-life deployment.

Yuhao Wu, Franziska Roesner, Tadayoshi Kohno et al.

Large language models (LLMs) extended as systems, such as ChatGPT, have begun supporting third-party applications. These LLM apps leverage the de facto natural language-based automated execution paradigm of LLMs: that is, apps and their interactions are defined in natural language, provided access to user data, and allowed to freely interact with each other and the system. These LLM app ecosystems resemble the settings of earlier computing platforms, where there was insufficient isolation between apps and the system. Because third-party apps may not be trustworthy, and exacerbated by the imprecision of natural language interfaces, the current designs pose security and privacy risks for users. In this paper, we evaluate whether these issues can be addressed through execution isolation and what that isolation might look like in the context of LLM-based systems, where there are arbitrary natural language-based interactions between system components, between LLM and apps, and between apps. To that end, we propose IsolateGPT, a design architecture that demonstrates the feasibility of execution isolation and provides a blueprint for implementing isolation, in LLM-based systems. We evaluate IsolateGPT against a number of attacks and demonstrate that it protects against many security, privacy, and safety issues that exist in non-isolated LLM-based systems, without any loss of functionality. The performance overhead incurred by IsolateGPT to improve security is under 30% for three-quarters of tested queries.