Cantilever double soldier-pile walls are used in vertical excavations to minimize wall deformation. A model is proposed consisting of an equivalent single soldier-pile wall with twice the bending stiffness of a single soldier-pile wall, twice the area of the pile shaft in contact with the soil below the excavation level, and the maximum capable mobilized moment at ground level, which can be used to evaluate the stability of double soldier-pile walls. A series of centrifuge model tests at an acceleration of 30 g was conducted to study the performance of cantilever single and double soldier-pile walls in sand under various test conditions. The test results show that the use of a double soldier pile can effectively improve the performance of retaining walls. The proposed stability analysis also provides satisfactory predictions of the factors of safety for double soldier-pile walls at various excavation depths and with various pile arrangements, verified by the results obtained in the centrifuge model tests. The use of F s ¼ 2.0 in the proposed stability analysis was found to be appropriate engineering practice for ensuring that the temporary cantilever double soldier-pile walls are stable and that the ratio hm /H does not exceed 1%.

Early detection increases overall survival among patients with lung cancer. This study formulated a machine learning method that processes chest X-rays (CXRs) to detect lung cancer early. After we preprocessed our dataset using monochrome and brightness correction, we used different kinds of preprocessing methods to enhance image contrast and then used U-net to perform lung segmentation. We used 559 CXRs with a single lung nodule labeled by experts to train a You Only Look Once version 4 (YOLOv4) deep-learning architecture to detect lung nodules. In a testing dataset of 100 CXRs from patients at Taipei Veterans General Hospital and 154 CXRs from the Japanese Society of Radiological Technology dataset, the sensitivity of the AI model using a combination of different preprocessing methods performed the best at 79%, with 3.04 false positives per image. We then tested the AI by using 383 sets of CXRs obtained in the past 5 years prior to lung cancer diagnoses. The median time from detection to diagnosis for radiologists assisted with AI was 46 (3–523) days, longer than that for radiologists (8 (0–263) days). The AI model can assist radiologists in the early detection of lung nodules.