Explore the vast range of titles available.

The prediction of anatomical structures within the surgical field by artificial intelligence (AI) is expected to support surgeons’ experience and cognitive skills. We aimed to develop a deep-learning model to automatically segment loose connective tissue fibers (LCTFs) that define a safe dissection plane. The annotation was performed on video frames capturing a robot-assisted gastrectomy performed by trained surgeons. A deep-learning model based on U-net was developed to output segmentation results. Twenty randomly sampled frames were provided to evaluate model performance by comparing Recall and F1/Dice scores with a ground truth and with a two-item questionnaire on sensitivity and misrecognition that was completed by 20 surgeons. The model produced high Recall scores (mean 0.606, maximum 0.861). Mean F1/Dice scores reached 0.549 (range 0.335–0.691), showing acceptable spatial overlap of the objects. Surgeon evaluators gave a mean sensitivity score of 3.52 (with 88.0% assigning the highest score of 4; range 2.45–3.95). The mean misrecognition score was a low 0.14 (range 0–0.7), indicating very few acknowledged over-detection failures. Thus, AI can be trained to predict fine, difficult-to-discern anatomical structures at a level convincing to expert surgeons. This technology may help reduce adverse events by determining safe dissection planes.

We aimed to develop an AI model that recognizes and displays loose connective tissue as a dissectable layer in real-time during gastrointestinal surgery and to evaluate its performance, including feasibility for clinical application. Training data were created under the supervision of gastrointestinal surgeons. Test images and videos were randomly sampled and model performance was evaluated visually by 10 external gastrointestinal surgeons. The mean Dice coefficient of the 50 images was 0.46. The AI model could detect at least 75% of the loose connective tissue in 91.8% of the images (459/500 responses). False positives were found for 52.6% of the images, but most were not judged significant enough to affect surgical judgment. When comparing the surgeon’s annotation with the AI prediction image, 5 surgeons judged the AI image was closer to their own recognition. When viewing the AI video and raw video side-by-side, surgeons judged that in 99% of the AI videos, visualization was improved and stress levels were acceptable when viewing the AI prediction display. The AI model developed demonstrated performance at a level approaching that of a gastrointestinal surgeon. Such visualization of a safe dissectable layer may help to reduce intraoperative recognition errors and surgical complications.

Laparoscopic hepatectomy (LH) requires accurate visualization and appropriate handling of hepatic veins and the Glissonean pedicle that suddenly appear during liver dissection. Failure to recognize these structures can cause injury, resulting in severe bleeding and bile leakage. This study aimed to develop a novel artificial intelligence (AI) system that assists in the visual recognition and color presentation of tubular structures to correct the recognition gap among surgeons.BACKGROUND/AIMLaparoscopic hepatectomy (LH) requires accurate visualization and appropriate handling of hepatic veins and the Glissonean pedicle that suddenly appear during liver dissection. Failure to recognize these structures can cause injury, resulting in severe bleeding and bile leakage. This study aimed to develop a novel artificial intelligence (AI) system that assists in the visual recognition and color presentation of tubular structures to correct the recognition gap among surgeons.Annotations were performed on over 350 video frames capturing LH, after which a deep learning model was developed. The performance of the AI was evaluated quantitatively using intersection over union (IoU) and Dice coefficients, as well as qualitatively using a two-item questionnaire on sensitivity and misrecognition completed by 10 hepatobiliary surgeons. The usefulness of AI in medical education was qualitatively evaluated by 10 medical students and residents.PATIENTS AND METHODSAnnotations were performed on over 350 video frames capturing LH, after which a deep learning model was developed. The performance of the AI was evaluated quantitatively using intersection over union (IoU) and Dice coefficients, as well as qualitatively using a two-item questionnaire on sensitivity and misrecognition completed by 10 hepatobiliary surgeons. The usefulness of AI in medical education was qualitatively evaluated by 10 medical students and residents.The AI model was able to individually recognize and colorize hepatic veins and the Glissonean pedicle in real time. The IoU and Dice coefficients were 0.42 and 0.53, respectively. Surgeons provided a mean sensitivity score of 4.24±0.89 (from 1 to 5; Excellent) and a mean misrecognition score of 0.12±0.33 (from 0 to 4; Fail). Medical students and residents assessed the AI to be very useful (mean usefulness score, 1.86±0.35; from 0 to 2; Excellent).RESULTSThe AI model was able to individually recognize and colorize hepatic veins and the Glissonean pedicle in real time. The IoU and Dice coefficients were 0.42 and 0.53, respectively. Surgeons provided a mean sensitivity score of 4.24±0.89 (from 1 to 5; Excellent) and a mean misrecognition score of 0.12±0.33 (from 0 to 4; Fail). Medical students and residents assessed the AI to be very useful (mean usefulness score, 1.86±0.35; from 0 to 2; Excellent).The novel AI presented was able to assist surgeons in the intraoperative recognition of microstructures and address the recognition gap among surgeons to ensure a safer and more accurate LH.CONCLUSIONThe novel AI presented was able to assist surgeons in the intraoperative recognition of microstructures and address the recognition gap among surgeons to ensure a safer and more accurate LH.

IntroductionWe retrospectively evaluated the blood coagulation activity using the D-dimer level in the early period after gastrectomy and investigated whether postoperative hypercoagulation affects tumor recurrence and long-term survival in gastric cancer patients.MethodsThe study involved 650 patients who underwent curative resection for gastric cancer at Kanagawa Cancer Center between July 2009 and July 2013. They were divided into a low-D-dimer group (LD group) and high-D-dimer group (HD group) according to the median D-dimer level on postoperative day (POD) 7. The risk factors for overall survival (OS) and relapse-free survival (RFS) were identified.ResultsOf the 448 enrolled patients, 218 were classified into the LD group and 230 into the HD group. The 5-year OS rates after surgery were 90.8% and 81.3% in the LD and HD groups, respectively (p < 0.001). The 5-year RFS rates after surgery were 89.9% and 76.1% in the LD and HD groups, respectively (p < 0.001). A high D-dimer level on POD 7 (≥ 4.9 μg/ml) was identified as an independent predictive factor for both the OS (hazard ratio [HR] 1.955, 95% confidence interval [CI] 1.158–3.303, p = 0.012) and RFS (HR 2.182, 95% CI 1.327–3.589, p = 0.002). Furthermore, hematological recurrence was significantly more frequent in the HD group than in the LD group (p = 0.014).ConclusionA high D-dimer level on POD 7 may predict tumor recurrence and the long-term survival in patients who undergo gastrectomy for locally advanced gastric cancer. Patients with an elevated postoperative D-dimer level need careful observation and diagnostic imaging to timely detect tumor recurrence.

BackgroundLaparoscopic gastrectomy is becoming more commonly performed, but acquisition of its technique remains challenging. We investigated whether laparoscopy-assisted distal gastrectomy (LDG) performed by trainees (TR) supervised by a technically qualified experienced surgeon (QS) is feasible and safe.MethodsThe short-term outcomes of LDG were assessed in patients with gastric cancer between 2008 and 2018. We compared patients who underwent LDG performed by qualified experienced surgeons (QS group) with patients who underwent LDG performed by the trainees (TR group).ResultsThe operation time was longer in the TR group than in the QS group (median time: 270 min vs. 239 min, p < 0.001). The median duration of the postoperative hospital stay was 9 days in the QS group and 8 days in the TR group (p = 0.003). The incidence of postoperative complications did not differ significantly between the two groups. Grade 2 or higher postoperative complications occurred in 18 patients (12.9%) in the QS group and 47 patients (11.7%) in the TR group (p = 0.763). Grade 3 or higher postoperative complications occurred in 9 patients (6.4%) in the QS group and 17 patients (4.2%) in the TR group (p = 0.357). Multivariate analysis showed that the American Society of Anesthesiologist Physical Status was an independent predictor of grade 2 or higher postoperative complications and that gender was an independent predictor of grade 3 or higher postoperative complications. The main operator (TR/QS) was not an independent predictor of complications.ConclusionsLaparoscopy-assisted distal gastrectomy performed by trainees supervised by an experienced surgeon is a feasible and safe procedure similar to that performed by experienced surgeons.

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. 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. 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. 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.

Background Surgery for gastric cancer should be performed as soon as possible after diagnosis. However, sometimes the waiting time for surgery tends to be longer. The relation between the waiting time for surgery and survival in patients with gastric cancer remains to be fully investigated. Methods This retrospective, single-center cohort study evaluated patients with gastric cancer who underwent curative surgery from 2006 through 2012 at Kanagawa Cancer Center in Japan. Patients who received neoadjuvant chemotherapy were excluded. The waiting time for surgery was defined as the time between the first visit and surgery. We investigated whether the waiting time for surgery has a linear negative impact on outcomes by using a Cox regression model with clinical prognostic factors. Results In total, 801 patients were eligible. The median waiting time was 45 days (range 10–269 days). The restricted cubic spline regression curve showed that the adjusted time-specific hazard ratios of waiting times did not indicate a linear negative trend on survival between 20 and 100 days ( p  = 0.759). In the Cox model with a quartile of waiting times, waiting times in the 32–44-day group, 43–62-day group, and ≥63 day groups were not associated with poorer overall survival as compared with the ≤31 day group (HR: 1.01, 95% CI 0.63–1.60, p  = 0.984, HR: 1.17, 95% CI 0.70–1.94, p  = 0.550, HR: 1.06, 95% CI 0.60–1.88, p  = 0.831, respectively). Conclusions There was no negative relation between the waiting time for surgery (within 100 days) and survival in patients with gastric cancer.

Background/Aim: Preoperative sarcopenia is associated with various cancers and affects the long-term prognosis of patients. After gastrectomy for gastric cancer, dynamic changes in body composition occur, and sarcopenia becomes more apparent after surgery than before surgery. However, the relationship between sarcopenia in the early postoperative period and long-term survival is not fully understood. The aim of this study was to determine the effects of surgical sarcopenia on long-term outcomes of patents with gastric cancer. Patients and Methods: We included 408 patients who underwent curative gastrectomy (distal or total gastrectomy) for gastric cancer at the Kanagawa Cancer Center from December 2013 to November 2017. Sarcopenia was defined using the skeletal muscle index (SMI), using computed tomography (CT) one month after gastrectomy. We compared the long-term outcomes between patients with and without sarcopenia. Results: The 5-year overall survival (OS) rates were 83.2% and 91.4% in the surgical and non-surgical sarcopenia groups, respectively. The hazard ratio (HR) of surgical sarcopenia for OS was 2.410 (95% confidence interval (CI)=1.321-4.396). In addition, surgical sarcopenia was associated with non-cancer-related deaths and deaths from other cancers. Conclusion: Patients with surgical sarcopenia after gastrectomy should be carefully monitored not only for gastric cancer recurrence but also for the occurrence of other diseases, including other cancers.

We hypothesised that the prognostic nutrition index (PNI) is useful for evaluating host immunity and response to immune checkpoint inhibitors. We investigated the effect of PNI on nivolumab monotherapy efficacy in advanced or recurrent gastric cancer (GC) or gastro-oesophageal junction cancer (GOC) patients. We retrospectively examined 110 patients, divided them into a high-PNI group and a low-PNI group, and compared treatment efficacy, adverse events (AEs), and survival between the groups. Median overall survival (OS) was significantly longer in the high-PNI group than in the low-PNI group (205 vs. 109 days; p<0.001). Multivariate analysis revealed that low PNI was an independent risk factor for OS (hazard ratio=2.398; 95% confidence interval=1.384-4.154; p=0.002). The overall response rate and frequency of AEs were not significantly different between the groups. PNI could be a useful prognostic factor in GC or GOC patients undergoing nivolumab monotherapy.

Background/Aim: The changes of dietary intake (DI) after gastrectomy have not been objectively reported. It has not been clear how much DI loss is experienced after total gastrectomy (TG) in comparison to after distal gastrectomy (DG). This study quantified the changes of DI after gastrectomy, and clarified how much DI loss is experienced after TG. Patients and Methods: This was a prospective observational study. Patients who underwent gastrectomy for gastric cancer were enrolled. The DI loss was evaluated at 1 and 3 months postoperatively. Results: Thirty-three patients underwent TG, and 117 patients underwent DG. The median %DI loss of the overall study population at 1 and 3 months after surgery was –9.3% and –3.6%. The median %DI loss at 1 and 3 months postoperatively was –15.6% and –5.3% in TG group, –8.9% and –3.3% in DG group (p=0.10 and 0.49, respectively). Conclusion: The patients experienced DI loss of approximately 10% at 1 month after gastrectomy. Patients who received TG tended to show a greater %DI loss at 1 month postoperatively.