The performance of deep learning on thyroid nodule imaging predicts thyroid cancer: A systematic review and meta-analysis of epidemiological studies with independent external test sets.

BACKGROUND AND AIMS: It is still controversial whether deep learning (DL) systems add accuracy to thyroid nodule imaging classification based on the recent available evidence. We conducted this study to analyze the current evidence of DL in thyroid nodule imaging diagnosis in both internal and external test sets. METHODS: Until the end of December 2022, PubMed, IEEE, Embase, Web of Science, and the Cochrane Library were searched. We included primary epidemiological studies using externally validated DL techniques in image-based thyroid nodule appraisal. This systematic review was registered on PROSPERO (CRD42022362892). RESULTS: We evaluated evidence from 17 primary epidemiological studies using externally validated DL techniques in image-based thyroid nodule appraisal. Fourteen studies were deemed eligible for meta-analysis. The pooled sensitivity, specificity, and area under the curve (AUC) of these DL algorithms were 0.89 (95% confidence interval 0.87-0.90), 0.84 (0.82-0.86), and 0.93 (0.91-0.95), respectively. For the internal validation set, the pooled sensitivity, specificity, and AUC were 0.91 (0.89-0.93), 0.88 (0.85-0.91), and 0.96 (0.93-0.97), respectively. In the external validation set, the pooled sensitivity, specificity, and AUC were 0.87 (0.85-0.89), 0.81 (0.77-0.83), and 0.91 (0.88-0.93), respectively. Notably, in subgroup analyses, DL algorithms still demonstrated exceptional diagnostic validity. CONCLUSIONS: Current evidence suggests DL-based imaging shows diagnostic performances comparable to clinicians for differentiating thyroid nodules in both the internal and external test sets.

Hepatitis C infection, related services, and barriers to HCV treatment among drug users in methadone maintenance treatment (MMT) clinics in Shanghai, China.

BACKGROUND: The purpose of this study was to document the prevalence of hepatitis C among MMT patients, hepatitis C virus (HCV) knowledge of patients and MMT staff members, and the barriers preventing them from receiving or delivering HCV-related services in MMT clinics of China. METHODS: Data were collected from 240 MMT patients and 58 staff members in Shanghai MMT clinics. Structured questionnaires (HCV Knowledge Scale and Alcohol Use Disorders Identification Test) and several self-developed questionnaires were used to assess (1) patient and staff HCV knowledge, (2) attitudes toward HCV-related services in MMT clinics, and (3) what type of HCV-related services the staff members have provided in their routine work. The HCV test results were based on the patients' medical records. RESULTS: The HCV seropositive rate was high (70%), and both patients and staff had limited HCV knowledge. The mean score of patient HCV knowledge was 6.8 out of 20 (SD = 3.7), whereas the mean score of staff HCV knowledge was 10.9 out of 20 (SD = 3.1). For HCV-positive patients, only 13.7% had accessed HCV medical treatment. Barriers included the cost of medical treatment, lack of HCV knowledge, lack of professional training for patients to receive HCV-related services from individuals or MMT clinics, and lack of an adequate policy-making system. CONCLUSIONS: HCV infection remains an important problem among MMT patients in China. Barriers to HCV-related services are attributable to individual, clinical, and policy-related factors. This study may provide evidence-based information for future work to optimize the resources of MMT clinics. TRIAL REGISTRATION: ClinicalTrials.gov NCT01647191 . Registered 17 April 2012.