RESUMEN
This study aimed to determine the value of either ultrasonography and computed tomography (CT) alone or both for diagnosing salivary gland stones. Based on their clinical findings and physical examination, 188 patients with salivary stones were included. Initially, an ultrasound was performed, and then a non-contrast-enhanced thin-sliced CT scan was done. The study included 161 patients with suspected submandibular gland (SMG) sialolithiasis and 27 with parotid gland (PG) sialolithiasis. Among the SMG cases, stones were confirmed in 130 patients through interventional sialoendoscopy. Sixteen of the stones were detected by CT scan only and were not seen in the previous ultrasound examination. After the second follow-up ultrasound, which was conducted after reviewing the CT scans, 9 of them were identified. Five patients with distal stones and ductal dilation on ultrasound were confirmed to have sialolithiasis. In 5 patients with stones detected on ultrasound but not on CT, dental filling artifacts were identified as the cause. For PG cases, stones were observed in 18 out of 19 patients with suspected sialolithiasis based on ultrasound and CT findings. Three patients with positive CT findings but negative ultrasound showed stones during sialendoscopy. During an ultrasound examination, palpating the floor of the mouth with the other hand's index finger can help identify stones obscured by the shadow of the mandible, thereby enhancing the test's accuracy. Patients with dental fillings, implants, and permanent dentures should not be referred for a CT scan to avoid unnecessary exposure to ionizing radiation. They also cause significant metallic artifacts in the field of interest. In these cases, ultrasound should be considered. To minimize the effects of ionizing radiation, it is recommended to create limited field requests for the target, focusing solely on the submandibular and PGs.
RESUMEN
PURPOSE: This study aims to evaluate the effectiveness of advanced deep learning models, specifically YOLOv8 and EfficientNetV2, in detecting meniscal tears on magnetic resonance imaging (MRI) using a relatively small data set. METHOD: Our data set consisted of MRI studies from 642 knees-two orthopaedic surgeons labelled and annotated the MR images. The training pipeline included MRI scans of these knees. It was divided into two stages: initially, a deep learning algorithm called YOLO was employed to identify the meniscus location, and subsequently, the EfficientNetV2 deep learning architecture was utilized to detect meniscal tears. A concise report indicating the location and detection of a torn meniscus is provided at the end. RESULT: The YOLOv8 model achieved mean average precision at 50% threshold (mAP@50) scores of 0.98 in the sagittal view and 0.985 in the coronal view. Similarly, the EfficientNetV2 model obtained area under the curve scores of 0.97 and 0.98 in the sagittal and coronal views, respectively. These outstanding results demonstrate exceptional performance in meniscus localization and tear detection. CONCLUSION: Despite a relatively small data set, state-of-the-art models like YOLOv8 and EfficientNetV2 yielded promising results. This artificial intelligence system enhances meniscal injury diagnosis by generating instant structured reports, facilitating faster image interpretation and reducing physician workload. LEVEL OF EVIDENCE: Level III.
RESUMEN
Inguinal hernia is the most common seen groin hernias which mostly contain bowel. The incidence of vermiform appendix in an inguinal hernia is seen in 1% of all inguinal hernia. This is known as Amyand's hernia. Appendix within a hernia can be normal or complicated by appendicitis. Most of these cases are not diagnosed preoperatively and managed during surgery. Preoperative diagnosis of these cases is so rare. Very few cases have been reported so far. In our case, we diagnosed an inflamed appendix in a 49 years old female within right inguinal hernia by using ultrasound and confirmed it by CT scan.