Introduction

Artificial Intelligence (AI) has significantly changed the landscape of healthcare, particularly in the diagnosis and treatment of Inflammatory Bowel Disease (IBD) and various digestive conditions. Key AI tools like ChatGPT and Google Gemini have the ability to decode endoscopic images, examine various samples, streamline administrative tasks, and aid in the evaluation of medical imagery and device automation. These AI solutions improve the management of digestive diseases by tailoring treatments and predicting potential adverse reactions. The incorporation of AI into healthcare allows patients to make informed decisions through the assessment of personalized treatment options.

Aims & Methods

The aim of the study was to assess the accuracy of two popular chatbots, ChatGPT and Google Gemini, in addressing medical management inquiries. Each chatbot was given a set of questions to answer, with their responses rated on a 1-10 Likert scale, where 1 represented a low score and 10 was high. To ensure fair evaluation, two neutral evaluators examined the responses of each bot. The research sought to highlight the abilities of these chatbots by methodically evaluating their performance in terms of accuracy and reliability. The use of two evaluators and the Likert scale method was instrumental in reducing potential biases, thereby validating the results.

Results

Our research evaluated the effectiveness of ChatGPT and Google Gemini in the realm of medical management. ChatGPT outperformed, attaining a 71% accuracy rate compared to Google BARD’s 39% (p = 0.032), and surpassed in information reliability with 58% against Google BARD’s 29% (p = 0.021). This study emphasizes the critical role of accuracy and reliability in the development of chatbots for managing IBS. ChatGPT's impressive performance suggests its potential as a reliable tool, yet it also points to the continuous need for research and improvement in this area.

Conclusion

AI has profoundly transformed healthcare, especially in the diagnosis and management of IBD. ChatGPT and Google Bard stand out as crucial AI instruments for interpreting medical imagery, evaluating samples, and enhancing efficiency. Our study assessed the accuracy and reliability of these chatbots within medical management contexts. ChatGPT outshone Google Bard by achieving 71% accuracy and a 58% reliability rate in information. ChatGPT demonstrates significant promise as a dependable AI tool for IBD care, yet our findings also underline the ongoing need for further research and development in the field.

References

1. Okagawa Y, Abe S, Yamada M, Oda I, Saito Y. Artificial Intelligence in Endoscopy. Dig Dis Sci. 2022 May;67(5):1553-1572. doi: 10.1007/s10620-021-07086-z. Epub 2021 Jun 21. PMID: 34155567.
2. Gomaa S, Posey J, Bashir B, Basu Mallick A, Vanderklok E, Schnoll M, Zhan T, Wen KY. Feasibility of a Text Messaging-Integrated and Chatbot-Interfaced Self-Management Program for Symptom Control in Patients With Gastrointestinal Cancer Undergoing Chemotherapy: Pilot Mixed Methods Study. JMIR Form Res. 2023 Nov 10;7:e46128. doi: 10.2196/46128. PMID: 37948108; PMCID: PMC10674151.

Introduction

Inflammatory bowel disease (IBD) represents a group of chronic inflammatory disorders of the gastrointestinal tract. While the hallmark features of IBD involve inflammation and damage to the digestive system, the impact of the disease often extends beyond the gut, manifesting as a spectrum of extraintestinal manifestations (EIM), affecting various organs and contributing to the complexity and heterogeneity of IBD.

Aims & Methods

The aim of this study was to assess the impact of EIM in IBD on disease course. We conducted a retrospective, descriptive and comparative study including patients hospitalized for IBD over a 9-year period [January 2011-January 2020]. Patients with less than 2 years of follow-up were excluded. Clinico-biological data, disease severity, and long-term outcomes (surgery and disease course) were collected. Data were entered and analyzed via SPSS 26 software.

Results

We included 225 men and 227 women. We enrolled 341 patients diagnosed with Crohn's disease (CD) and 111 patients with ulcerative colitis (UC). The mean age at diagnosis was 36.15 ± 12.74 years. One-hundred and sixty-five patients (36.4%) had EIM with respectively: rheumatological manifestations (n=144), dermatological manifestations (n=26), ocular manifestations (n=17) and hepatic manifestations (n=8). We noticed a significant association of EIM with male gender (56.3% vs 45.9%; p=0.03). Rheumatological EIM were associated with a significant older age at disease onset (36.7 years vs 32.8 years; p=0.003). No significant association was noted between EIM with neither family history of IBD nor the type of IBD (p=0.33 and p=0.3 respectively). In case of CD, we noted a higher frequency of ileal involvement in patients with EIM, without a significant difference (61.4% vs 51.4%; p=0.07). No association was noted between EIM with neither fistulising behaviour nor with extended ileal disease nor with upper gastrointestinal tract involvement (p=0.1; p=0.5 and p=0.5 respectively). In case of UC, no association was noted between EIM and an extensive disease (p=0.7). As for treatment, EIM had no impact on corticosteroids’ use or immunosuppressive treatment (p=0.36 and p= 0.08), but was associated with a significant higher use of anti-TNF therapy (83.4% vs 27.2%; p=0.01). Response to corticosteroids and anti-TNF therapy was similar whether with or without EIM (p= 0.36 and p=0.24 respectively). In the sub-group of ocular EIM, we noted a higher failure of immunosuppressive treatment (70% vs 36.6%; p=0.04). We noted that EIM was associated with a rarer urge to surgical treatment (35.7% vs 44.9%;p=0.05) with a significant longer timeframe to surgery (47.7 months vs 27,2 months; p=0.03). The mean number of years of active disease was significantly higher in patients with EIM (3.3 vs 2.5; p=0.002) with a higher number of flare-ups (p=0.009) and hospitalizations (p<0.001), reflecting a better disease course in patients without EIM.

Conclusion

In our study, EIM were associated with an overall more severe disease course in IBD, thus, emphasizing the importance of comprehensive assessment and management strategies tailored to address both intestinal and extra-intestinal manifestations in the holistic care of IBD patients.

Introduction

Inflammatory bowel diseases (IBD) carry a high risk of complications if not diagnosed promptly and are often misdiagnosed as irritable bowel syndrome (IBS). Delayed IBD detection leads to an increase in complicated cases and missed opportunities for timely therapy initiation.

Aims & Methods

The study aimed to develop an AI-based clinical decision support system (CDSS) for IBD diagnosis.
We created three databases, containing clinical-laboratory data, endoscopic and histopathological images. Patients enrolled from 2014 to 2024 included those with confirmed IBD (ulcerative colitis (UC) and Crohn’s disease (CD) with colonic involvement) during flare-ups and IBS with normal endoscopic and histopathological findings.
Separate artificial neural networks were designed for IBD differentiation: multilayer perceptrons (MLPs) for clinical-laboratory data analysis and a pre-trained VGG16 convolutional neural network (CNN) for digital image analysis (initialized with ImageNet weights).
This work has been supported by the grants the Russian Science Foundation, RSF 24-25-20127.

Results

A retrospective database of 1,406 patients with clinical and laboratory data was compiled. For IBD vs. IBS differentiation, the cohort was reduced to 390 patients that had all the fulfilled data, while UC vs. CD analysis included 964 patients.
An MLP (13:13-5-1:1) achieved 92.7% and 99.0% accuracy in IBD diagnosis, incorporating stool frequency/consistency, absence of extraintestinal/perianal manifestations, functional disorders, family history of autoimmune diseases, prior surgeries, CRP, and fecal calprotectin (p≤0.019).
Another MLP (9:9-8-1:1) detected 70.16% of CD and 86.40% of UC cases, using data on constipation, diarrhea, bloody stools, absence of extraintestinal manifestations, fever, perianal lesions, strictures, and prior bowel surgeries (p<0.001).
The endoscopic database comprised 5,693 JPEG images: 3,400 of normal mucosa, 1,357 - CD, 935 - UC. Model sensitivity and specificity were for CD - 91%/79%; UC - 90%/97%; normal - 90%/97%.
The histopathological dataset included 398 whole-slide images (1280×960 px), segmented into 224×224 px tiles. The final set included 8,620 images: 2,360 - CD, 2,320 – normal mucosa, 3,950 - UC. Sensitivity and specificity were for CD - 91% and 95%, UC - 96%, 95%, normal - 96%, 93%, respectively.
The integrated clinical DSS was constructed using Bayesian belief networks implemented in Hugin Lite 8.3 software and integrated clinical, endoscopic, and histopathological nodes. The network architecture replicated the stepwise diagnostic process and incorporated the following nodes: clinical and laboratory data, endoscopy (IBD/IBS differentiation), histopathology (IBD/IBS differentiation), IBS, IBD, endoscopy (UC/CD diagnosis), histopathology (UC/CD discrimination), final diagnosis. The system was validated using six patients per diagnostic category (IBS, UC, CD), with complete sets of clinical, endoscopic, and histopathological data required for all component models. The integrated model demonstrated an overall diagnostic accuracy of 83.3%.

Conclusion

The developed CDSS shows promise for earlier IBD detection but requires further validation studies.

Introduction

The enhancement of the quality of life (QoL) of patients with inflammatory bowel disease (IBD) has become a major therapeutic target in the management of this disease. Sexuality is one of its main determinants. However, in Tunisia, there is little data on sexual dysfunction (SD) and fertility among IBD patients.

Aims & Methods

The aim of this study was to determine the incidence and the factors associated with SD during IBD. We conducted a cross-sectional study including patients with IBD followed at the Gastroenterology Department of Charles Nicole Hospital between July and December 2020. Sexual function was assessed by the Female Sexual Function Index (FSFI) score in women and by the International Index of Erectile Function (IIEF) in men. Scores assessing patients ‘overall QoL were calculated: Hospital anxiety and depression scale (HADS), Fatigue Severity Scale (FSS), and Short Inflammatory Bowel Disease Questionnaire (S-IBDQ). All patients gave oral informed consent before answering the questionnaires.

Results

A total of 65 patients were included of whom 72.3% had CD and 27.7% had UC, with a mean age of 47.58±10.18 years and sex ratio (M/F) of 0.44.
SD was found in 73.3% of women according to the FSFI and in 15% of men according to the IIEF-15. Erectile dysfunction (ED) was observed in 40% of patients based on the IIEF-5.
Female gender was a significant factor of SD (p<0.001; OR 15.6; CI [3.9-62.8]).
An analytical statistical study was conducted, showing factors associated with SD in the female patient group only.
Our study showed that the factors associated with the presence of female SD were age (p=0.018), absence of professional activity (p=0.034), length of marriage (p=0.022), presence of probable or overt depression (p=0.01), fatigue (p=0.042), impaired QoL judged by the S-IBDQ (p=0, 042), the four domains of the S-IBDQ (bowel symptoms (p=0.04), general signs (p=0.011), emotional domain (p=0.004), social domain (p=0.013)), IBD-disk score (p=0.007), and IBD-disk domains (sexual activity (p<0.001), emotion (p=0.001), energy (p=0.016), and joint pain (p=0.015)).
There were no factors that appeared to be associated with male SD. Meanwhile, we found a statistically significant association between men's perception of impaired sexuality due to IBD and the presence of ED (p=0.004; OR=33; CI [2.4-443.6]).
In multivariate analysis, the factors independently associated with female SD were the duration of marriage (p=0.018; OR 3.58; CI [1.43-7.83]), the presence of suspected or overt depression (p<0.001; OR 12.6; CI [1.82-72.52]) and fatigue (p=0.032; OR 11.8; CI [1.61-96.7]).

Conclusion

Our study shows that sexual disorders are frequent in IBD patients, especially in women. A multidisciplinary management involving gastroenterologists, psychologists, and sexologists is sometimes necessary in these young patients, in order to improve their QoL.

Introduction

Inflammatory bowel diseases (IBD) are systemic disorders that may be associated with extraintestinal manifestations in up to 40% of patients. Recognizing the recurrent emergence of these manifestations in IBD, with a particular emphasis on rheumatological aspect, becomes a pivotal focal point for a holistic therapeutic approach.

Aims & Methods

The aim of this study was to evaluate the impact of ankylosing spondylitis in IBD on disease course. We conducted a retrospective and comparative study, including patients followed for IBD over an 8-year period [2014 - 2022]. Clinico-biological data, disease severity, and long-term outcomes (surgery and disease course) were collected. IBD patients with ankylosing spondylitis (group1) and patients without (group2) were compared. Data were entered and analyzed via SPSS 26 software.

Results

One-hundred IBD patients were enrolled with a sex ratio of M/F=2.44 and a mean age of disease diagnose of 37.26 ± 13.46 years. Eighteen patients had ankylosing spondylitis (18%). The two groups were comparable in terms of age at disease onset, gender, smoking status and family history of IBD (p=0.65; p=0.9, p=1 and 0.1 respectively). In Crohn's disease, there was no difference between the two groups regarding the extent of the disease (p=0.71), the presence of upper gastro-intestinal tract involvement (p=0.11) and perineal manifestations (p=0.7). Inflammatory, structuring, and penetrating phenotypes were also comparable between the two groups (p=0.67; p=0.3; p=1 respectively). However, the occurrence of acute intestinal obstruction was significantly more common in ankylosing spondylitis patients (G1:54% vs. G2:10%; p=0.004). As for ulcerative colitis, extensive presentation (G1:28% vs. G2:41%; p=0.66) and the occurrence of disease extension during follow-up was comparable between the two groups (p=0.78). None of the patients with or without rheumatological involvement had any dermatological or ophthalmological involvement related to IBD, but there was a more significant prevalence of hepatic manifestations in ankylosing spondylitis group (G1:16% vs. G2:2%, p=0.01). The mean number of flare-ups was not significantly different between the two groups (G1:3.21±1.47 vs. G2:2.53±1.4; p=0.06). Regarding treatment, recourse to methotrexate (G1:16% vs. G2:2%; p=0.01) and the use of anti-TNF agents (G1:77% vs. G2:22%; p<0.001) was more frequent in group 1, with no difference in the occurrence of adverse events (p=1), the use of corticosteroids (p=0.22) or azathioprine (p=0.1). Surgical treatment was more frequently indicated in ankylosing spondylitis patients (G1:51% vs. G2:30%; p=0.06). Post-operative relapse of Crohn's disease was comparable between the two groups (p=1).

Conclusion

In our study, rheumatological involvement was not likely responsible of a more severe clinical phenotype of IBD.

Introduction

Intestinal Ultrasound (IUS) is an accurate, non-invasive technique to monitor disease activity and guide treatment decisions in inflammatory bowel disease (IBD). It has gained popularity as a point-of-care (POC) modality due to its tolerability and ease-of-use and reproducibility of results.

Aims & Methods

This study aims to assess the impact of POC IUS in daily clinical management of IBD. We conducted a single-centre retrospective cohort study of IBD patients who were evaluated with POC IUS between September 2022 and March 2023. Ultrasound data were retrieved including ultrasound reports, static images. and cine loop with doppler activity. Clinical data were retrieved from the electronic medical record. Descriptive statistics were used to evaluate characteristics of the study population. Chi-squared tests and logistic regression were used for categorical outcomes.

Results

A total of 210 IUS examinations were included in patients with Crohn’s disease (n=118, 56%), ulcerative colitis (n=90, 43%) and IBD-unclassified (n=2, 1%). The median age was 35.6years and the median disease duration was 7.0years. Fifty percent of IUS (n=105) was performed on asymptomatic patients as part of routine care; twenty-five percent (n=53) was for symptomatic patients; and twenty-three percent (n=49) was a follow-up imaging to assess response to a change in therapy. A change in management following IUS was recorded in 117 cases (55.7%).
For routine examinations (n=105), 22.9% (n=24) were abnormal, seven (7%) were referred for urgent endoscopy, thirty-five cases (33.3%) changed therapy, twenty (19%) initiated new therapy and four (4%) optimised existing therapy. Among symptomatic patients who underwent IUS (n=53), a change of medication was documented in 89% (n=47), endoscopy performed after examinations in 15% (n=8), 23% (n=12) were prescribed laxatives and 2% (n=1) were admitted to hospital.
Abnormal IUS and change in medication following IUS was higher in symptomatic than asymptomatic patients (54.7% vs 23.1%, p<0.001 and 54.7% vs 12.0%, p<0.001, respectively). Faecal calprotectin (n=105) (p=0.03) and magnetic resonance imaging (n=8) (p=0.06) results within a month prior to IUS were directly correlated with bowel wall thickness >3mm on IUS. Nineteen endoscopies were done within 1 month after examinations, which was also positively correlated with IUS result (p=0.01).

Conclusion

Our real-world study showed that POC IUS plays a role in ‘treat-to-target’ strategy in patient-management decisions, especially among symptomatic patients, allowing rapid treatment decisions. In addition to other objective markers of disease, including faecal calprotectin, POC IUS has the potential to reduce the need for more invasive endoscopic assessment of disease activity.

Disclosure

None

Introduction

Bariatric surgery emerges as a powerful intervention for obesity, promoting enhanced metabolic health. However, substantial evidence indicates a rise in Irritable Bowel Syndrome (IBS), primarily following such surgical interventions. The effect of race on the occurrence of IBS after surgery has yet to be investigated. Given the widespread occurrence of IBS and its confirmed connection to bariatric surgery, examining the effect of racial disparities is crucial. This study seeks to ascertain whether racial differences impact the probability of an individual developing IBS post-bariatric surgery. Revealing this relationship is vital for improving patient results and optimizing postoperative care.

Aims & Methods

We executed a retrospective study to analyze patients who underwent bariatric surgery at our facility between 2009 and 2022. Data about comorbidities, insurance statuses, surgical techniques, colonoscopy and pathology reports, and initial patient attributes were collected through ICD and CPT codes. Patients who developed IBS post-surgery were monitored until 2023 and categorized by gender, race, and duration post-surgery. We used propensity score matching to balance baseline characteristics. Participants with a previous diagnosis of IBS were excluded from the study. Kaplan-Meier survival analysis and Cox proportional hazards models were leveraged to examine the onset of IBS among different racial groups. Additionally, odds ratios were employed to discern independent variables influencing the study outcomes

Results

Between 2009 and 2022, 960 bariatric surgeries were performed at our institution. Of these, 39 patients, equating to 0.04%, were diagnosed with IBS, typically around 40±9.2 months post-surgery. The subjects' average age was approximately 39±8.7 years, with females constituting 51.2% of the participants. Regarding racial breakdown, 29.2% were African Americans, 33.5% were Whites, 21.8% were Hispanics, and 15.5% were Asians. Notably, White patients manifested symptoms of IBS significantly earlier compared to other racial groups (23±5vs 36±7, 31±8, 29±6) (Breslow: 12.46, p=0.015). Additionally, a heightened risk of IBS was observed among individuals with lower educational attainment (OR: 1.25, P=0.025).

Conclusion

This research aims to bridge the prevailing knowledge void relating to the influence of race on the emergence of IBS following bariatric surgery. Our retrospective analysis of 960 surgeries from 2009 to 2022 highlighted a 0.04% occurrence of IBS, generally identified around 40 months after surgery. Notably, White patients experienced the onset of IBS in considerably less time than other racial groups, with individuals with lower educational levels exhibiting increased risks. These insights emphasize the importance of considering racial and other sociodemographic elements in refining postoperative care

References

1. Turner PL, Oyetunji TA, Gantt G, Chang DC, Cornwell EE, Fullum TM. Demographically associated variations in outcomes after bariatric surgery. Am J Surg. 2011 Apr;201(4):475-80. doi: 10.1016/j.amjsurg.2010.08.035. PMID: 21421101.