Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 2 appraisals
BMJ open
Protocol for evaluating the effects of the intelligent multimodal symptom assessment and response system in patients with pemphigus vulgaris: a randomised controlled trial
INTRODUCTION: Pemphigus vulgaris is a rare autoimmune blistering disease characterised by recurrent mucocutaneous erosions, high symptom burden and unpredictable relapse. Current management relies mainly on pharmacological therapy and hospital-based follow-up, with limited real-time monitoring and individualised support in home-based disease management. To address these challenges, this trial aims to evaluate the effectiveness of an Intelligent Multimodal Symptom Assessment and Response System, integrating patient-reported outcomes, wearable physiological data and image-based lesion assessments, to improve symptom management and quality of life in pemphigus vulgaris patients. Primary objective is to evaluate the effectiveness of the Intelligent Multimodal Symptom Assessment and Response System in improving symptom alerting and symptom management outcomes in patients with pemphigus vulgaris. Secondary objectives are to enhance patients' self-management ability in symptom monitoring and control, to improve treatment adherence throughout the follow-up period, to promote health-related quality of life among pemphigus vulgaris patients, to assess the usability and acceptability of the system from the patients' perspective. METHODS AND ANALYSIS: This is a multicentre, parallel-group, randomised controlled trial. 160 participants will be randomly assigned to either the intervention group (receiving the Intelligent Multimodal Symptom Assessment and Response System) or the control group (receiving standard care). Eligible participants will be adults aged 18 years or older with a confirmed diagnosis of pemphigus vulgaris and active skin or mucosal lesions, who are able to use digital devices and provide written informed consent. Individuals with severe comorbidities, concurrent participation in other clinical trials or cognitive impairments that may interfere with study adherence will be excluded. The intervention will be delivered via a digital platform, integrating electronic patient-reported outcomes, wearable physiological data and lesion images over a 12-week follow-up period. ETHICS AND DISSEMINATION: Ethical approval was obtained from the Institutional Review Board of the School of Nursing, Fudan University (IRB 2025-07-13), the Medical Ethics Committee of the Institute of Dermatology, Chinese Academy of Medical Sciences (2025-KY-034) and the Ethics Committee of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine (2025-452). Approvals were granted on 10 September, 18 July and 23 July 2025, respectively. This protocol is based on V1.0, 13 July 2025 of the protocol. The results of this study will be disseminated through peer-reviewed publications and academic conferences. TRIAL REGISTRATION NUMBER: ChiCTR2500109711.
29 May 2026
Read appraisal →IEEE transactions on medical imaging
Self-Paced Learning for Images of Antinuclear Antibodies
Antinuclear antibody (ANA) testing is a critical method for diagnosing autoimmune disorders such as Lupus, Sjögren's syndrome, and scleroderma. Despite its importance, manual ANA detection is slow, labor-intensive, and demands years of training. ANA detection is complicated by over 100 coexisting antibody types, resulting in vast fluorescent pattern combinations. Although machine learning and deep learning have enabled automation, ANA detection in real-world clinical settings presents unique challenges as it involves multi-instance, multi-label (MIML) learning. In this paper, a novel framework for ANA detection is proposed that handles the complexities of MIML tasks using unaltered microscope images without manual preprocessing. Inspired by human labeling logic, it identifies consistent ANA sub-regions and assigns aggregated labels accordingly. These steps are implemented using three task-specific components: an instance sampler, a probabilistic pseudo-label dispatcher, and self-paced weight learning rate coefficients. The instance sampler suppresses low-confidence instances by modeling pattern confidence, while the dispatcher adaptively assigns labels based on instance distinguishability. Self-paced learning adjusts training according to empirical label observations. Our framework overcomes limitations of traditional MIML methods and supports end-to-end optimization. Extensive experiments on one ANA dataset and three public medical MIML benchmarks demonstrate the superiority of our framework. On the ANA dataset, our model achieves up to +7.0% F1-Macro and +12.6% mAP gains over the best prior method, setting new state-of-the-art results. It also ranks top-2 across all key metrics on public datasets, reducing Hamming loss and one-error by up to 18.2% and 26.9%, respectively. The source code can be accessed at https://github.com/fletcherjiang/ANA-SelfPacedLearning.
4 Apr 2026
Read appraisal →