CT-based deep learning segmentation of ovarian cancer and the stability of the extracted radiomics features.

Background: Radiomics analysis could provide complementary tissue characterization in ovarian cancer (OC). However, OC segmentation required in radiomics analysis is time-consuming and labour-intensive. In this study, we aim to evaluate the performance of deep learning-based segmentation of OC on contrast-enhanced CT images and the stability of radiomics features extracted from the automated segmentation. Methods: Staging abdominopelvic CT images of 367 patients with OC were retrospectively recruited. The training and cross-validation sets came from center A (n=283), and testing set (n=84) came from centers B and C. The tumours were manually delineated by a board-certified radiologist. Four model architectures provided by no-new-Net (nnU-Net) method were tested in this task. The segmentation performance evaluated by Dice score, Jaccard score, sensitivity and precision were compared among 4 architectures. The Pearson correlation coefficient (ρ), concordance correlation coefficient (ρc) and Bland-Altman plots were used to evaluate the volumetric assessment of OC between manual and automated segmentations. The stability of extracted radiomics features was evaluated by intraclass correlation coefficient (ICC). Results: The 3D U-Net cascade architecture achieved highest median Dice score, Jaccard score, sensitivity and precision for OC segmentation in the testing set, 0.941, 0.890, 0.973 and 0.925, respectively. Tumour volumes of manual and automated segmentations were highly correlated (ρ=0.944 and ρc =0.933). 85.0% of radiomics features had high correlation with ICC >0.8. Conclusions: The presented deep-learning segmentation could provide highly accurate automated segmentation of OC on CT images with high stability of the extracted radiomics features, showing the potential as a batch-processing segmentation tool.

Machine learning application for the prediction of SARS-CoV-2 infection using blood tests and chest radiograph.

Triaging and prioritising patients for RT-PCR test had been essential in the management of COVID-19 in resource-scarce countries. In this study, we applied machine learning (ML) to the task of detection of SARS-CoV-2 infection using basic laboratory markers. We performed the statistical analysis and trained an ML model on a retrospective cohort of 5148 patients from 24 hospitals in Hong Kong to classify COVID-19 and other aetiology of pneumonia. We validated the model on three temporal validation sets from different waves of infection in Hong Kong. For predicting SARS-CoV-2 infection, the ML model achieved high AUCs and specificity but low sensitivity in all three validation sets (AUC: 89.9-95.8%; Sensitivity: 55.5-77.8%; Specificity: 91.5-98.3%). When used in adjunction with radiologist interpretations of chest radiographs, the sensitivity was over 90% while keeping moderate specificity. Our study showed that machine learning model based on readily available laboratory markers could achieve high accuracy in predicting SARS-CoV-2 infection.

Efficacy and safety of ultrasound-guided high-intensity focused ultrasound for uterine fibroids: a preliminary experience.

The objective of this prospective cohort study was to evaluate the therapeutic efficacy and safety of ultrasound-guided high-intensity focussed ultrasound (HIFU) in the treatment of uterine fibroids. Twenty premenopausal women with symptomatic fibroids underwent ultrasound-guided HIFU therapy. Twenty-two fibroids with a median baseline volume of 127.0 cm3 (range 18.5-481.2 cm3) were treated. The percentages fibroid volume reduction were 46.9 (range -8.8-73.1) at 1-month, 57.4 (-51.5-95.2) at 3-month, 60.1 (-18.9-97.8) at 6-month and 75.9 (-33.7-99.3) at 12-month, after treatment. The modified Uterine Fibroid Symptom and Quality of Life (UFS-QOL) scores were reduced by 40.7% (0-59.3%) at 3-month, 45.5% (0-70.4%) at 6-month and 44.9% (0-71.4%) at 12-month after treatment. Three patients required subsequent surgical interventions. No significant complications were encountered. Ultrasound-guided HIFU appears to be effective and safe for the treatment of symptomatic uterine fibroids in selected patients. Impact statement What is already known on this subject? Ultrasound-guided high-intensity focussed ultrasound (HIFU) is a relatively new uterine-sparing treatment for fibroids. Most clinical reports are from China, which suggest that this treatment is a safe and effective modality. However, in many other countries, HIFU treatment for fibroids, especially using ultrasound as image guidance, is still considered novel with limited clinical experience. What do the results of this study add? This preliminary report adds to our limited local experience on HIFU and provides reassurance on our continual utilisation of this treatment modality for fibroids. With the increasing demand of uterine-sparing alternatives, experiences shared among different countries are important to make this treatment modality generalisable and universally acceptable. What are the implications of these findings for clinical practice and/or further research?Ultrasound-guided HIFU (USgHIFU) can potential be offered as an alternative treatment modality for women with fibroids.

Ovarian Reserve After Ultrasound-Guided High-Intensity Focused Ultrasound for Uterine Fibroids: Preliminary Experience.

OBJECTIVE: To determine the effect on ovarian reserve of ultrasound-guided high-intensity focused ultrasound (HIFU) in the treatment of uterine fibroids. METHODS: We performed a mid-study analysis of markers of ovarian reserve using data from a prospective cohort study evaluating the safety of ultrasound-guided HIFU for uterine fibroids. Blood samples obtained from 12 women with uterine fibroids less than one week before treatment were used for measurement of serum anti-Mullerian hormone (AMH), and this testing was repeated in the first, third, sixth, and 12th month after ultrasound-guided HIFU treatment. RESULTS: Fourteen fibroids from 12 patients were treated using ultrasound-guided HIFU. The median baseline fibroid volume was 101.2 cm(3) (range 18.5 to 349.2 cm(3)). The median treatment time was 140.5 minutes (46 to 192 minutes), and median sonication time was 1449 seconds (range 541 to 2445 seconds). The median energy delivered was 575 521.5 joules (range 216 400 to 898 273 joules). The median AMH levels (ng/mL) before treatment and at one, three, six, and 12 months after treatment were 0.3 (range 0.01 to 1.94), 0.47 (0.01 to 1.43), 0.205 (0.01 to 1.81), 0.26 (0 to 2.37), and 0.06 (0.02 to 1.04), respectively. There was no significant difference between the AMH levels before and at any time after treatment. No patient became amenorrheic or reported symptoms suggestive of menopause after treatment. CONCLUSION: Our preliminary experience suggests that ovarian reserve does not seem to be affected by ultrasound-guided HIFU in the treatment of uterine fibroids.