Background Histotripsy is a nonthermal, nonionizing, noninvasive, focused US technique that relies on cavitation for mechanical tissue breakdown at the focal point. Preclinical data have shown its safety and technical success in the ablation of liver tumors. Purpose To evaluate the safety and technical success of histotripsy in destroying primary or metastatic liver tumors. Materials and Methods The parallel United States and European Union and England #HOPE4LIVER trials were prospective, multicenter, single-arm studies. Eligible patients were recruited at 14 sites in Europe and the United States from January 2021 to July 2022. Up to three tumors smaller than 3 cm in size could be treated. CT or MRI and clinic visits were performed at 1 week or less preprocedure, at index-procedure, 36 hours or less postprocedure, and 30 days postprocedure. There were co-primary end points of technical success of tumor treatment and absence of procedure-related major complications within 30 days, with performance goals of greater than 70% and less than 25%, respectively. A two-sided 95% Wilson score CI was derived for each end point. Results Forty-four participants (21 from the United States, 23 from the European Union or England; 22 female participants, 22 male participants; mean age, 64 years ± 12 [SD]) with 49 tumors were enrolled and treated. Eighteen participants (41%) had hepatocellular carcinoma and 26 (59%) had non-hepatocellular carcinoma liver metastases. The maximum pretreatment tumor diameter was 1.5 cm ± 0.6 and the maximum post-histotripsy treatment zone diameter was 3.6 cm ± 1.4. Technical success was observed in 42 of 44 treated tumors (95%; 95% CI: 84, 100) and procedure-related major complications were reported in three of 44 participants (7%; 95% CI: 2, 18), both meeting the performance goal. Conclusion The #HOPE4LIVER trials met the co-primary end-point performance goals for technical success and the absence of procedure-related major complications, supporting early clinical adoption. Clinical trial registration nos. NCT04572633, NCT04573881 Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Nezami and Georgiades in this issue.
Background It is increasingly recognized that interstitial lung abnormalities (ILAs) detected at CT have potential clinical implications, but automated identification of ILAs has not yet been fully established. Purpose To develop and test automated ILA probability prediction models using machine learning techniques on CT images. Materials and Methods This secondary analysis of a retrospective study included CT scans from patients in the Boston Lung Cancer Study collected between February 2004 and June 2017. Visual assessment of ILAs by two radiologists and a pulmonologist served as the ground truth. Automated ILA probability prediction models were developed that used a stepwise approach involving section inference and case inference models. The section inference model produced an ILA probability for each CT section, and the case inference model integrated these probabilities to generate the case-level ILA probability. For indeterminate sections and cases, both two- and three-label methods were evaluated. For the case inference model, we tested three machine learning classifiers (support vector machine [SVM], random forest [RF], and convolutional neural network [CNN]). Receiver operating characteristic analysis was performed to calculate the area under the receiver operating characteristic curve (AUC). Results A total of 1382 CT scans (mean patient age, 67 years ± 11 [SD]; 759 women) were included. Of the 1382 CT scans, 104 (8%) were assessed as having ILA, 492 (36%) as indeterminate for ILA, and 786 (57%) as without ILA according to ground-truth labeling. The cohort was divided into a training set (n = 96; ILA, n = 48), a validation set (n = 24; ILA, n = 12), and a test set (n = 1262; ILA, n = 44). Among the models evaluated (two- and three-label section inference models; two- and three-label SVM, RF, and CNN case inference models), the model using the three-label method in the section inference model and the two-label method and RF in the case inference model achieved the highest AUC, at 0.87. Conclusion The model demonstrated substantial performance in estimating ILA probability, indicating its potential utility in clinical settings. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Zagurovskaya in this issue.