Stroke, which includes cerebral infarction, cerebral hemorrhage, and subarachnoid hemorrhage, is the fourth leading cause of death among Japanese people. Conquering a stroke is a significant challenge as it often affects patients, their families, and social resources. When medical treatment is ineffective in preventing recurrence, surgical intervention becomes the preferred option. Surgical treatment for stroke lesions can be done in two ways: direct surgery, which involves making a skin incision and a craniotomy to reach the lesion, and endovascular surgery, which 0515 involves using a catheter delivered via the radial or femoral artery to the lesion. During endovascular surgery, coils, embolic materials, and/or stents may be used to treat the lesion. Minimally invasive endovascular neurosurgery is generally preferred by patients, and its usage has increased over the years. However, some lesions may be more effectively treated with direct surgery rather than endovascular treatment. Imaging plays a critical role in the treatment strategy. Preoperative imaging is used to determine whether the lesion can be treated by endovascular surgery. Intraoperative imaging helps in selecting an appropriate device during treatment, and postoperative imaging is used to determine if recurrence or re-treatment is necessary. In recent years, various diagnostic imaging devices such as DSA (digital subtraction angiography), MRI (magnetic resonance imaging) and CT (computed tomography) have been developed, offering improved image quality and providing less invasive procedures for patients. This paper will present the latest perioperative imaging methods for endovascular surgery, including carotid artery stenosis, acute ischemic stroke, and unruptured aneurysms, using state-of-the-art imaging modalities.
Ultrasonography is inexpensive and noninvasive compared to CT or MRI, but its limited range of motion, frequent artifacts, and dependence on operator experience make it very difficult for radiologists to accurately determine the criticality of chronic kidney disease (CKD). Therefore, the purpose of this study was to use deep learning to classify CKD of categories in ultrasound images with high accuracy. We used the right and left kidney long-axis tomogram, right and left kidney short-axis tomogram of 547 cases (G1: 75 cases, G2: 263 cases, G3: 186 cases, G4+5: 23 cases) who underwent renal ultrasound examination. We investigated seven deep learning models: AlexNet, GoogLeNet, ResNet18, ResNet50, ResNet101, Xception, and DenseNet201 with and without pretraining. Among the seven deep learning models, Xception with pretraining had the highest Area Under the Receiver Operating Characteristic Curve (AUROC): G1: 0.64±0.11, G2: 0.62±0.05, G3: 0.65±0.03, G4+G5: 0.60±0.08. In addition, the accuracy for each category in the test data was 65.1%. Since the number of cases in G1, G4, and G5 is very small compared to the number of cases in G2 and G3, we believe it is necessary to increase the number of training data by inverting or rotating the data vertically or horizontally in the future.