RESUMO
Ovarian cancer is the sixth most common malignancy, with a 35% survival rate across all stages at 10 years. Ultrasound is widely used for ovarian tumour diagnosis, and accurate pre-operative diagnosis is essential for appropriate patient management. Artificial intelligence is an emerging field within gynaecology and has been shown to aid in the ultrasound diagnosis of ovarian cancers. For this study, Embase and MEDLINE databases were searched, and all original clinical studies that used artificial intelligence in ultrasound examinations for the diagnosis of ovarian malignancies were screened. Studies using histopathological findings as the standard were included. The diagnostic performance of each study was analysed, and all the diagnostic performances were pooled and assessed. The initial search identified 3726 papers, of which 63 were suitable for abstract screening. Fourteen studies that used artificial intelligence in ultrasound diagnoses of ovarian malignancies and had histopathological findings as a standard were included in the final analysis, each of which had different sample sizes and used different methods; these studies examined a combined total of 15,358 ultrasound images. The overall sensitivity was 81% (95% CI, 0.80-0.82), and specificity was 92% (95% CI, 0.92-0.93), indicating that artificial intelligence demonstrates good performance in ultrasound diagnoses of ovarian cancer. Further prospective work is required to further validate AI for its use in clinical practice.
RESUMO
In this study, we aimed to systematicallyreview the current evidence regarding the diagnostic accuracy of ultrasound in assessing adnexal masses in pregnancy. The Cochrane Register of Controlled Trials, PubMed, and EMBASE databases were searched for all types of clinical studies that utilised ultrasound for the diagnosis of adnexal masses in pregnancy. Only studies that used outcome measures of either histological diagnosis or significant regression of the adnexal mass on imaging follow-up were included. The quality of each study was assessed for risk of bias. The diagnostic performance of ultrasound in each study type was calculated, along with the pooled diagnostic performance of ultrasound in differentiating benign from malignant masses. The initial search yielded 4,915 articles, of which 2,547 qualified for abstract screening. A total of 83 articles were included in this review, including one prospective cohort study, six retrospective observational studies, seven case series, and 69 case reports. In the included studies, the total number of adnexal masses was 559. The mean patient age was 29.2 years (95% confidence interval [CI]: 28.7-29.7), with a mean gestational age at diagnosis of 13.8 weeks (95% CI: 13.2-14.4). The mean quality assessment score was 75%. The International Ovarian Tumour Analysis Simple Rules were used in two articles, whereas subjective impression was used in the remaining 81 articles. The most frequently diagnosed mass was a simple or physiological cyst (35%). The prevalence of malignancy in the entire sample was 46/559 (8%; 95% CI: 34-61%). The overall pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of ultrasound in detecting ovarian malignancy were 64% (95% CI: 30-88%), 88% (95% CI: 64-97%), 5.6 (95% CI: 1.2-25.4), and 0.4 (95% CI: 0.15-1), respectively. In conclusion, currently, there is a lack of high-quality prospective studies to guide the management of adnexal masses in pregnancy. Ultrasound appears to have an adequate accuracy in differentiating benign from malignant masses; however, more research is required to assess the role of ultrasound models, rules, and subjective assessment in pregnancy compared to non-pregnant women.
