Daniel Singer
Research Subjects
Daniel Singer has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Ailsa Hart
Division of Digestive Diseases, Department of Metabolism, Digestion and Reproduction, Faculty of Medicine, Imperial College London , London
2 shared publications - Anna Antoniou
St Mark’s The National Bowel Hospital, London NW10 7NS, UK
2 shared publications - Christopher Tomlinson
University College London
2 shared publications - Easan Anand
Imperial College London London UK
2 shared publications - Gita Lingam
St Mark's Hospital and Academic Institute London UK
2 shared publications - Itai Ghersin
Department of Gastroenterology, Rambam Health Care Campus, Haifa 3109601, Israel
2 shared publications
Latest Publications
- Open Access
- 25 December 2025
- Published by MDPI AG in Diagnostics
- Vol. 16 (1) , 72
Enhancing Patient Understanding of Perianal Fistula MRI Findings Using ChatGPT: A Randomized, Single Centre Study
Background/Objectives: Large Language Models (LLMs) may help translate complex Magnetic Resonance Imaging (MRI) fistula reports into accessible, patient-friendly summaries. This study evaluated the clinical utility, safety, and patient acceptability of Generative Pre-trained Transformer (GPT-4o) in generating such reports. Methods: A three-phase study was conducted at a single centre. Phase I involved prompt engineering and pilot testing of GPT-4o outputs for feasibility. Phase II assessed 250 consecutive MRI fistula reports from September 2024 to November 2024, each reviewed by a multi-disciplinary panel to determine hallucinations and thematic content. Phase III randomised patients to review either a simple or complex fistula case, each containing an original report and an Artificial Intelligence (AI)-generated summary (order randomised, origin blinded), and rate readability, trustworthiness, usefulness and comprehension. Results: Sixteen patients participated in Phase I pilot testing. In Phase II, hallucinations occurred in 11% of outputs, with unverified recommendations also identified. In Phase III, 61 patients (mean age 48, 41% female) evaluated paired original and AI-generated summaries. AI summaries scored significantly higher for readability, comprehension, and usefulness than original reports (all p < 0.001), with equivalent trust ratings. Mean Flesch-Kincaid scores were markedly higher for AI-generated summaries (66 vs. 26; p < 0.001). Clinicians highlighted improved anatomical structuring and accessible language, but emphasised risks of inaccuracies. A revised template incorporating Multi-Disciplinary Team (MDT)-focused action points and a lay summary section was co-developed. Conclusions: LLMs can enhance the readability and patient understanding of complex MRI reports but remain limited by hallucinations and inconsistent terminology. Safe implementation requires structured oversight, domain-specific refinement, and clinician validation. Future development should prioritise standardised reporting templates incorporating clinician-approved lay summaries.
- 25 December 2025
- Published by MDPI AG in Diagnostics
- Vol. 16 (1) , 72
Open AccessEnhancing Patient Understanding of Perianal Fistula MRI Findings Using ChatGPT: A Randomized, Single Centre Study
Background/Objectives: Large Language Models (LLMs) may help translate complex Magnetic Resonance Imaging (MRI) fistula reports into accessible, patient-friendly summaries. This study evaluated the clinical utility, safety, and patient acceptability of Generative Pre-trained Transformer (GPT-4o) in generating such reports. Methods: A three-phase study was conducted at a single centre. Phase I involved prompt engineering and pilot testing of GPT-4o outputs for feasibility. Phase II assessed 250 consecutive MRI fistula reports from September 2024 to November 2024, each reviewed by a multi-disciplinary panel to determine hallucinations and thematic content. Phase III randomised patients to review either a simple or complex fistula case, each containing an original report and an Artificial Intelligence (AI)-generated summary (order randomised, origin blinded), and rate readability, trustworthiness, usefulness and comprehension. Results: Sixteen patients participated in Phase I pilot testing. In Phase II, hallucinations occurred in 11% of outputs, with unverified recommendations also identified. In Phase III, 61 patients (mean age 48, 41% female) evaluated paired original and AI-generated summaries. AI summaries scored significantly higher for readability, comprehension, and usefulness than original reports (all p < 0.001), with equivalent trust ratings. Mean Flesch-Kincaid scores were markedly higher for AI-generated summaries (66 vs. 26; p < 0.001). Clinicians highlighted improved anatomical structuring and accessible language, but emphasised risks of inaccuracies. A revised template incorporating Multi-Disciplinary Team (MDT)-focused action points and a lay summary section was co-developed. Conclusions: LLMs can enhance the readability and patient understanding of complex MRI reports but remain limited by hallucinations and inconsistent terminology. Safe implementation requires structured oversight, domain-specific refinement, and clinician validation. Future development should prioritise standardised reporting templates incorporating clinician-approved lay summaries.
- Open Access
- 4 September 2025
- Published by MDPI AG in Journal of Imaging
- Vol. 11 (9) , 302
AI-Generated Patient-Friendly MRI Fistula Summaries: A Pilot Randomised Study
Perianal fistulising Crohn’s disease (pfCD) affects 1 in 5 Crohn’s patients and requires frequent MRI monitoring. Standard radiology reports are written for clinicians using technical language often inaccessible to patients, which can cause anxiety and hinder engagement. This study evaluates the feasibility and safety of AI-generated patient-friendly MRI fistula summaries to improve patient understanding and shared decision-making. MRI fistula reports spanning healed to complex disease were identified and used to generate AI patient-friendly summaries via ChatGPT-4. Six de-identified MRI reports and corresponding AI summaries were assessed by clinicians for hallucinations and readability (Flesch-Kincaid score). Sixteen patients with perianal fistulas were randomized to review either AI summaries or original reports and rated them on readability, comprehensibility, utility, quality, follow-up questions, and trustworthiness using Likert scales. Patients rated AI summaries significantly higher in readability (median 5 vs. 2, p = 0.011), comprehensibility (5 vs. 2, p = 0.007), utility (5 vs. 3, p = 0.014), and overall quality (4.5 vs. 4, p = 0.013), with fewer follow-up questions (3 vs. 4, p = 0.018). Clinicians found AI summaries more readable (mean Flesch-Kincaid 54.6 vs. 32.2, p = 0.005) and free of hallucinations. No clinically significant inaccuracies were identified. AI-generated patient-friendly MRI summaries have potential to enhance patient communication and clinical workflow in pfCD. Larger studies are needed to validate clinical utility, hallucination rates, and acceptability.
- 4 September 2025
- Published by MDPI AG in Journal of Imaging
- Vol. 11 (9) , 302
Open AccessAI-Generated Patient-Friendly MRI Fistula Summaries: A Pilot Randomised Study
Perianal fistulising Crohn’s disease (pfCD) affects 1 in 5 Crohn’s patients and requires frequent MRI monitoring. Standard radiology reports are written for clinicians using technical language often inaccessible to patients, which can cause anxiety and hinder engagement. This study evaluates the feasibility and safety of AI-generated patient-friendly MRI fistula summaries to improve patient understanding and shared decision-making. MRI fistula reports spanning healed to complex disease were identified and used to generate AI patient-friendly summaries via ChatGPT-4. Six de-identified MRI reports and corresponding AI summaries were assessed by clinicians for hallucinations and readability (Flesch-Kincaid score). Sixteen patients with perianal fistulas were randomized to review either AI summaries or original reports and rated them on readability, comprehensibility, utility, quality, follow-up questions, and trustworthiness using Likert scales. Patients rated AI summaries significantly higher in readability (median 5 vs. 2, p = 0.011), comprehensibility (5 vs. 2, p = 0.007), utility (5 vs. 3, p = 0.014), and overall quality (4.5 vs. 4, p = 0.013), with fewer follow-up questions (3 vs. 4, p = 0.018). Clinicians found AI summaries more readable (mean Flesch-Kincaid 54.6 vs. 32.2, p = 0.005) and free of hallucinations. No clinically significant inaccuracies were identified. AI-generated patient-friendly MRI summaries have potential to enhance patient communication and clinical workflow in pfCD. Larger studies are needed to validate clinical utility, hallucination rates, and acceptability.