Peptides DB
Research-centric peptide and protocol reference hub
Study 2 of 7Evuzamitide literatureJournal of obstetrics and gynaecology : the journal of the Institute of Obstetrics and Gynaecology · Observational2026

Evaluating AI in postoperative rehabilitation of axillary web syndrome after breast cancer surgery: ChatGPT vs DeepSeek.

DeepSeek outperformed ChatGPT in rehabilitation decision support for axillary web syndrome, but findings should be interpreted with caution due to the limited case number.

Read at Journal of obstetrics and gynaecology : the journal of the Institute of Obstetrics and GynaecologyAdd to compare

Where it sits

this study against the rest of the evuzamitide corpus
0
Preclinical
4
Observational · this one
0
Open-label
1
Randomised
2
Reviews

Summary and findings

This study compared the performance of two AI models, ChatGPT and DeepSeek, in the rehabilitation management of axillary web syndrome after breast cancer surgery. Five virtual cases were evaluated by three rehabilitation specialists using a standardized scoring rubric. DeepSeek showed superior performance in several assessment dimensions compared to ChatGPT.

How much of this paper we could read: full text read (0.80). We had a clear abstract, so the summary below closely tracks the paper. What this means →
DeepSeek performed significantly better than ChatGPT in diagnostic accuracy (p=0.006).2026

Abstract

The authors’ words, as Journal of obstetrics and gynaecology : the journal of the Institute of Obstetrics and Gynaecology supplied them

<h4>Objective</h4>To systematically compare two generative artificial intelligence (AI) models, ChatGPT and DeepSeek in the rehabilitation management for axillary web syndrome (AWS) following breast cancer surgery; to assess their potential value in supporting clinical rehabilitation decision-making.<h4>Methods</h4>Five virtual cases, based on real clinical scenarios and representing varying levels of complexity and functional impairment, were constructed. Three rehabilitation medicine specialists independently evaluated the models' responses using a standardised scoring rubric. Assessment dimensions included diagnostic accuracy, comprehensiveness of evaluation, individualisation of treatment planning, safety and risk awareness, evidence-based support, and clarity of logic. Additional analyses examined clinical reasoning depth, error patterns, and differences in response style.<h4>Results</h4>Both models demonstrated high inter-rater reliability. For overall clinical competency scores, DeepSeek performed significantly better than ChatGPT. DeepSeek showed superior performance in diagnostic accuracy (<i>P</i> = 0.006), comprehensiveness of evaluation (<i>P</i> = 0.034), and individualisation of therapeutic planning (<i>P</i> = 0.005). DeepSeek maintained robust performance across cases of varying complexity, with particularly strong advantages in diagnostically challenging cases and those involving concomitant lymphedema. ChatGPT more frequently exhibited generalised treatment recommendations and omissions in differential diagnoses, whereas DeepSeek, despite occasionally provided recommendations with lower levels of evidence, had a lower overall error rate. DeepSeek generated more actionable and context-specific responses, often including concrete exercise prescriptions and technical details. It also outperformed ChatGPT in reasoning chain completeness, depth of evidence justification, and mechanistic explanation rationality.<h4>Conclusion</h4>In this exploratory pilot study based on a limited number of virtual cases, DeepSeek demonstrates relatively better performance than ChatGPT in AWS rehabilitation decision support. However, findings should be interpreted cautiously, and further validation in real-world clinical settings is required.

Elsewhere in the Evuzamitide corpus

BSupporting change in a complex reality: a qualitative study of specialist nurses' perceptions of counselling patients at risk of type 2 diabetes.International journal of nursing studies advances · 2026HumanCMimic antibodies: leveraging ligand mimicry for epitope-targeted antibody discovery.mAbs · 2026 · 11 of 31 candidates bound the IL-18RA D3 domain, with eight achieving sub-nanomolar affinities.In vitroDTraining and supervision of lay mental health workers in community-based interventions in East Africa: a scoping review.Global health action · 2026 · Not reported in abstract.reviewBEarly and mature &lt;i&gt;Achromobacter xylosoxidans&lt;/i&gt; biofilm in cystic fibrosis and non-cystic fibrosis isolates: dynamics and response to clinically relevant antibiotics.Biofilm · 2026 · Not reported in abstract.HumanBLongitudinal symptom dynamics in postoperative patients with liver cancer: A cross-lagged panel network analysis.Asia-Pacific journal of oncology nursing · 2026 · Jaundice predicted vomiting (OR = 1.94, 95% CI [1.70, 2.14]) from T1 to T2.HumanADesign and rationale of the Research with EVuzamitide to Evaluate for cardiac AmyLoidosis (REVEAL) positron emission tomography study: A study of (124)I-evuzamitide to diagnose cardiac amyloidosis.Journal of nuclear cardiology : official publication of the American Society of Nuclear Cardiology · 2026 · Not reported in abstract.Human