Precise highlighting of the pancreas by semantic segmentation during robot-assisted gastrectomy: visual assistance with artificial intelligence for surgeons.

BACKGROUND: A postoperative pancreatic fistula (POPF) is a critical complication of radical gastrectomy for gastric cancer, mainly because surgeons occasionally misrecognize the pancreas and fat during lymphadenectomy. Therefore, this study aimed to develop an artificial intelligence (AI) system capable of identifying and highlighting the pancreas during robot-assisted gastrectomy. METHODS: A pancreas recognition algorithm was developed using HRNet, with 926 training images and 232 validation images extracted from 62 scenes of robot-assisted gastrectomy videos. During quantitative evaluation, the precision, recall, intersection over union (IoU), and Dice coefficients were calculated based on the surgeons' ground truth and the AI-inferred image from 80 test images. During the qualitative evaluation, 10 surgeons answered two questions related to sensitivity and similarity for assessing clinical usefulness. RESULTS: The precision, recall, IoU, and Dice coefficients were 0.70, 0.59, 0.46, and 0.61, respectively. Regarding sensitivity, the average score for pancreas recognition by AI was 4.18 out of 5 points (1 = lowest recognition [less than 50%]; 5 = highest recognition [more than 90%]). Regarding similarity, only 54% of the AI-inferred images were correctly differentiated from the ground truth. CONCLUSIONS: Our surgical AI system precisely highlighted the pancreas during robot-assisted gastrectomy at a level that was convincing to surgeons. This technology may prevent misrecognition of the pancreas by surgeons, thus leading to fewer POPFs.

Treatment strategy for successful conversion surgery in clinical stage IVB gastric cancer.

INTRODUCTION: Recent advances in chemotherapy have resulted in successful conversion surgery (CS) for clinical stage (cStage) IVB gastric cancer (GC). This study aimed to evaluate the success rate of CS in clinical practice and determine optimal treatment strategies. METHODS: Totally, 166 patients with cStage IVB gastric and gastroesophageal junction adenocarcinoma, who underwent chemotherapy at Hyogo Medical University Hospital between January 2017 and June 2022, were included. CS was performed after confirming tumor to be M0 based on imaging and/or staging laparoscopy, except for resectable liver metastases. Preoperative chemotherapy was continued for at least 6 months provided that adverse events were manageable. RESULTS: Of 125 eligible patients, 23 were treated with CS, achieving a conversion rate of 18.4% and an R0 resection rate of 91.3%. The median duration of preoperative chemotherapy was 8.5 months; the median number of cycles was eight. The highest conversion rate was observed in patients receiving first-line treatment (14.4%), followed by those receiving second and third lines (5.8% and 2.3%, respectively). The median survival time in patients who received CS was significantly longer than that in patients who continued chemotherapy alone (56.7 versus 16 months, respectively, P < 0.0001). There was no significant difference in the 3-year overall survival between the patients who achieved CS after first-line treatment (63.2%, n = 18) and those who achieved CS after second- or third-line treatment (66.7%, n = 5). CONCLUSION: Consistent chemotherapy strategies could lead to successful CS and improved prognosis in a greater number of patients with cStage IVB GC, regardless of line of treatment.

[A Case of Breast Cancer in the Elderly Localized in the Nipple].

An 89-year-old woman presented to our clinic with a complaint of a wound in the left nipple. The pathological diagnosis via skin biopsy revealed adenocarcinoma. A physical examination, mammography, ultrasonography, and CT scan revealed a mass in the left nipple. However, a definite diagnosis could not be established by skin biopsy. Left breast-conserving surgery and sentinel lymph node biopsy were performed for diagnosis and treatment. Histological examination revealed a scirrhous type invasive ductal carcinoma in the left nipple and areola with skin invasion and lymph node metastasis. Radiation and hormone therapy were used as adjuvant therapy. She is alive with no local recurrence for 5 months post surgery. Since ductal carcinoma of the nipple is uncommon, we present this case report along with a review of the relevant literature.

Laparoscopic and left thoracoscopic Ivor-Lewis esophagectomy for Siewert type II esophagogastric junction cancer with right aortic arch: a case report.

BACKGROUND: Right aortic arch (RAA) is a congenital malformation detected in 0.04% of the population without heterotaxia and makes esophagectomy and mediastinal lymphadenectomy difficult. A left thoracic approach is recommended in patients with RAA, but a minimally invasive procedure has not yet been established. CASE PRESENTATION: The case was a 40-year-old man with RAA and Siewert type II adenocarcinoma of the esophagogastric junction with metastases to the adrenal glands and paraaortic lymph nodes. Conversion surgery was performed when radiologic disappearance of metastatic disease was confirmed after first-line treatment consisting of 12 cycles of S-1 plus platinum-based systemic chemotherapy. Minimally invasive laparoscopic and left thoracoscopic Ivor-Lewis esophagectomy was performed in the right semi-lateral decubitus position. The esophagus was easy to see on left thoracoscopy because of the RAA. Esophagectomy with lower mediastinal lymphadenectomy and an intrathoracic esophagogastric anastomosis was performed successfully with laparoscopy and thoracoscopy without a position change. There were no surgical complications, and no residual cancer was detected in the resected specimen on pathological examination. There has been no recurrence during 21 months of follow-up. CONCLUSIONS: Laparoscopic and left thoracoscopic Ivor-Lewis esophagectomy in the right semi-lateral decubitus position is a minimally invasive, anatomically novel procedure for Siewert type II esophagogastric junction cancer in patients with RAA.