To characterise 3-year pelvic pain and urinary symptom trajectories and to identify baseline factors associated with urologic chronic pelvic pain syndrome (UCPPS) improvement.
The Trans-Multidisciplinary Approach to the Study of Chronic Pelvic Pain (MAPP) Symptom Patterns Study was a multicentre, prospective cohort study of UCPPS, including interstitial cystitis/bladder pain syndrome and chronic prostatitis/chronic pelvic pain syndrome. Patients completed four weekly run-in assessments, baseline visit, and quarterly visits up to 3 years, providing clinical and patient-reported data. A functional clustering approach, applied separately to Pelvic Pain Severity (PPS) and Urinary Symptom Severity (USS) longitudinal change scores, was used to generate symptom trajectory clusters dichotomised as Group 0 ‘improvers’ vs Groups 1–3 ‘non-improvers’. Logistic regression models explored baseline factors associated with improvement and included run-in period average and baseline scores to adjust for regression to the mean effects.
A total of 545 patients (66% female) were followed for a median (interquartile range) of 34 (23–35) months. Four trajectory clusters were identified for each of PPS and USS, consistent with moderate improvement (Group 0), slight improvement (Group 1), no change (Group 2), and slight worsening (Group 3). In all, 18% and 19% of patients were in the moderately improved PPS and USS groups, respectively, representing 30% of patients overall. Female sex, better sleep, and less opioid use were associated with PPS improvement (Group 0); younger age and baseline cystoscopic treatment were associated with USS improvement (Group 0).
In all, 30% of patients with UCPPS demonstrated improvement in pain and/or urinary symptoms over 3 years. Baseline factors associated with improvement may represent markers of a milder or localised phenotype and/or treatment effects.
To develop a deep learning model for differential diagnosis of acute scrotum using single ultrasound (US) images.
We analysed 1172 patients with acute scrotal pain evaluated by Doppler US at four hospitals. From each case, we selected a representative axial colour Doppler US image. We trained a binary classification model to distinguish torsion from non-torsion using an EfficientNet architecture. The dataset was split 70% for training and 30% for validation. We addressed class imbalance with data augmentation and class weighting. Class Activation Mapping was used to interpret model decisions.
The model achieved robust performance: accuracy 97%, precision 98%, sensitivity 97%, and F1 score 97%. Class activation mapping heatmaps localised decision-making to pathologically critical regions, including absent testicular blood flow and whirlpool signs. In a 20-patient prospective pilot study, the system correctly identified both surgically confirmed torsion cases, with one non-torsion case misclassified as torsion.
A deep learning model demonstrated promising diagnostic performance in differentiating acute scrotal emergencies using single US images. Its feasibility was preliminarily assessed in a small pilot study. These findings support further investigation, with larger and more balanced multicentre studies required to establish clinical utility and effective workflow integration.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：