Accuracy of thoracic nerves recognition for surgical support system using artificial intelligence.

We developed a surgical support system that visualises important microanatomies using artificial intelligence (AI). This study evaluated its accuracy in recognising the thoracic nerves during lung cancer surgery. Recognition models were created with deep learning using images precisely annotated for nerves. Computational evaluation was performed using the Dice index and the Jaccard index. Four general thoracic surgeons evaluated the accuracy of nerve recognition. Further, the differences in time lag, image quality and smoothness of movement between the AI system and surgical monitor were assessed. Ratings were made using a five-point scale. The computational evaluation was relatively favourable, with a Dice index of 0.56 and a Jaccard index of 0.39. The AI system was used for 10 thoracoscopic surgeries for lung cancer. The accuracy of thoracic nerve recognition was satisfactory, with a recall score of 4.5 ± 0.4 and a precision score of 4.0 ± 0.9. Though smoothness of motion (3.2 ± 0.4) differed slightly, nearly no difference in time lag (4.9 ± 0.3) and image quality (4.6 ± 0.5) between the AI system and the surgical monitor were observed. In conclusion, the AI surgical support system has a satisfactory accuracy in recognising the thoracic nerves.

An artificial intelligence-based nerve recognition model is useful as surgical support technology and as an educational tool in laparoscopic and robot-assisted rectal cancer surgery.

BACKGROUND: Artificial intelligence (AI) has the potential to enhance surgical practice by predicting anatomical structures within the surgical field, thereby supporting surgeons' experiences and cognitive skills. Preserving and utilising nerves as critical guiding structures is paramount in rectal cancer surgery. Hence, we developed a deep learning model based on U-Net to automatically segment nerves. METHODS: The model performance was evaluated using 60 randomly selected frames, and the Dice and Intersection over Union (IoU) scores were quantitatively assessed by comparing them with ground truth data. Additionally, a questionnaire was administered to five colorectal surgeons to gauge the extent of underdetection, overdetection, and the practical utility of the model in rectal cancer surgery. Furthermore, we conducted an educational assessment of non-colorectal surgeons, trainees, physicians, and medical students. We evaluated their ability to recognise nerves in mesorectal dissection scenes, scored them on a 12-point scale, and examined the score changes before and after exposure to the AI analysis videos. RESULTS: The mean Dice and IoU scores for the 60 test frames were 0.442 (range 0.0465-0.639) and 0.292 (range 0.0238-0.469), respectively. The colorectal surgeons revealed an under-detection score of 0.80 (± 0.47), an over-detection score of 0.58 (± 0.41), and a usefulness evaluation score of 3.38 (± 0.43). The nerve recognition scores of non-colorectal surgeons, rotating residents, and medical students significantly improved by simply watching the AI nerve recognition videos for 1 min. Notably, medical students showed a more substantial increase in nerve recognition scores when exposed to AI nerve analysis videos than when exposed to traditional lectures on nerves. CONCLUSIONS: In laparoscopic and robot-assisted rectal cancer surgeries, the AI-based nerve recognition model achieved satisfactory recognition levels for expert surgeons and demonstrated effectiveness in educating junior surgeons and medical students on nerve recognition.

Laparoscopic distal gastrectomy skill evaluation from video: a new artificial intelligence-based instrument identification system.

The advent of Artificial Intelligence (AI)-based object detection technology has made identification of position coordinates of surgical instruments from videos possible. This study aimed to find kinematic differences by surgical skill level. An AI algorithm was developed to identify X and Y coordinates of surgical instrument tips accurately from video. Kinematic analysis including fluctuation analysis was performed on 18 laparoscopic distal gastrectomy videos from three expert and three novice surgeons (3 videos/surgeon, 11.6 h, 1,254,010 frames). Analysis showed the expert surgeon cohort moved more efficiently and regularly, with significantly less operation time and total travel distance. Instrument tip movement did not differ in velocity, acceleration, or jerk between skill levels. The evaluation index of fluctuation ß was significantly higher in experts. ROC curve cutoff value at 1.4 determined sensitivity and specificity of 77.8% for experts and novices. Despite the small sample, this study suggests AI-based object detection with fluctuation analysis is promising because skill evaluation can be calculated in real time with potential for peri-operational evaluation.

Carbon ion radiotherapy for mesonephric adenocarcinoma of the uterine cervix: a case report.

BACKGROUND: Mesonephric adenocarcinoma is an extremely rare subtype of uterine cervical cancer that is associated with a poor prognosis and for which a standardized treatment protocol has not been established. Carbon ion radiotherapy (CIRT) is an emerging radiotherapy modality that has been shown to have a favorable anti-tumor effect, even for tumors resistant to conventional photon radiotherapy or chemotherapy. However, there is no report on CIRT outcomes for mesonephric adenocarcinoma of the uterine cervix. CASE PRESENTATION: We treated a 47-year-old Japanese woman with mesonephric adenocarcinoma of the uterine cervix (T2bN0M0 and stage IIB according to the 7th edition of the Union for International Cancer Control and International Federation of Gynecology and Obstetrics, respectively) with CIRT combined with brachytherapy and concurrent chemotherapy. CIRT consisted of whole pelvic irradiation and boost irradiation to the gross tumor; 36.0 Gy (relative biological effectiveness [RBE]) in 12 fractions and 19.2 Gy (RBE) in 4 fractions, respectively, performed once a day, four times per week. Computed tomography-based image-guided adaptive brachytherapy was performed after completion of CIRT, for which the D90 (i.e., the dose prescribed to 90% of the target volume) for the high-risk clinical target volume was 20.4 Gy in a total of 3 sessions in 2 weeks. A weekly cisplatin (40 mg/m2) dose was administered concomitantly with the radiotherapy for a total of five courses. From 4 months post-CIRT, the patient developed metastasis of the lung, with a total of 10 lung metastases over 70 months; these lesions were treated on each occasion by photon stereotactic body radiotherapy and/or systemic therapy. At 8 years from initial treatment (i.e., 2 years after the last treatment), the patient is alive without any evidence of recurrence and maintains a high quality of life. CONCLUSIONS: This is the first report of CIRT for treatment of mesonephric adenocarcinoma of the uterine cervix. The present case indicates the potential efficacy of CIRT in combination with brachytherapy for treatment of this disease.

Novel navigation for laparoscopic cholecystectomy fusing artificial intelligence and indocyanine green fluorescent imaging.

This preliminary study is the first to demonstrate that AI can precisely identify loose connective tissue during laparoscopic cholecystectomy and ICG fluorescent cholangiography. Tashiro and colleagues conclude that this novel real-time navigation modality fusing AI and ICG fluorescent imaging may enhance safety and provide more reliable laparoscopic or robotic surgery.

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.

Color-coded laparoscopic liver resection using artificial intelligence: A preliminary study.

Tashiro and colleagues demonstrated for the first time that an artificial intelligence system can precisely identify intrahepatic vascular structures during laparoscopic liver resection in real time through color coding under bleeding and indocyanine green fluorescent imaging. The system supports real-time navigation and offers potentially safer laparoscopic or robotic liver surgery.

Development of a Novel Artificial Intelligence System for Laparoscopic Hepatectomy.

BACKGROUND/AIM: 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. PATIENTS AND METHODS: 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. RESULTS: 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). CONCLUSION: 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.

[A Case of Advanced Esophagogastric Junction Adenocarcinoma with Enlarged Lymph Node Metastases of the Middle Mediastinum and Intraperitoneal That Exhibited pCR by Preoperative SOX Therapy].

A 79-year-old man was diagnosed with esophagogastric junction adenocarcinoma, cT3N3M0, cStage Ⅲ, including enlarged lymph node metastases(Bulky N)in the middle mediastinum and intraperitoneal. A total of 2 cycles of S-1 plus oxaliplatin(SOX)was administered. After neoadjuvant chemotherapy, the primary tumor and enlarged lymph nodes had greatly decreased in size. Subsequently, thoracoscopic subtotal esophagectomy and reconstruction with a gastric tube were performed. Histopathological examinations showed no residual cancer cells in the primary lesion and dissected lymph nodes (pathological complete response). Preoperative chemotherapy containing SOX could be a useful treatment strategy for patients with esophagogastric junction adenocarcinoma with enlarged lymph node metastasis.

Reducing the risk of postoperative pancreatic fistula in radical gastrectomy: pre-assessment with computed tomography for the diagnosis of pancreatic steatosis.

PURPOSE: To determine whether pancreatic steatosis (PS) is associated with the risk of postoperative pancreatic fistula (POPF) after radical gastrectomy, and if so, to investigate whether pre-assessment by diagnostic imaging can mitigate the risk. METHODS: The clinical records of 276 patients with cStage I gastric cancer who underwent laparoscopic gastrectomy with D1 + lymphadenectomy between 2012 and 2015 were reviewed. In the first phase up to July 2013 (n = 138), PS was classified from computed tomography (CT) findings into type S (superficial fat deposition) or type D (diffuse fatty replacement) and examined for association with POPF. In the second phase (n = 138), the preoperative CT assessment of PS was routinized. Separate samples from pancreatoduodenectomy consistent with each type were histologically examined. RESULTS: In the first phase, the incidence of POPF was significantly higher in group S, but not in group D, compared with normal pancreas (16.3% and 9.1% vs. 3.6%, respectively; P = 0.03). The drain amylase level was lowest in group D, reflecting exocrine insufficiency. Histologically, the loose connective-tissue space between the fat infiltrating the pancreas and the peripancreatic fat containing the lymph nodes was unclear in type D but conserved in type S. In the second phase, surgery was performed with more intention on accurately tracing the dissection plane and significantly lowered incidence of POPF in Group S (16.3% to 2.1%; P = 0.047). CONCLUSION: Peripancreatic lymphadenectomy is more challenging and likely to cause POPF in patients with PS. However, the risk may be reduced using appropriate dissection techniques based on the CT pre-assessment findings.

Automated segmentation by deep learning of loose connective tissue fibers to define safe dissection planes in robot-assisted gastrectomy.

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.

Colorectal Cancer Screening Outcomes of 2412 Prostate Cancer Patients Considered for Carbon Ion Radiotherapy.

Colorectal cancer (CRC) screening is effective for detecting cancer in average-risk adults. For prostate cancer (PCa) patients considered for carbon ion radiotherapy (CIRT), pre-treatment CRC screening is performed empirically to avoid post-treatment colonoscopic manipulation. However, the outcomes of screening this population remain unclear. Here, we compared the outcomes of routine pre-CIRT CRC screening of 2412 PCa patients at average risk for CRC with data from two published datasets: the Japan National Cancer Registry (JNCR) and a series of 17 large-scale screening studies analyzing average-risk adults. The estimated prevalence rate was calculated using the pooled sensitivity elucidated by a previous meta-analysis. Consequently, 28 patients (1.16%) were diagnosed with CRC. CRC morbidity was significantly associated with high pre-treatment levels of prostate-specific antigen (p = 0.023). The screening positivity rate in this study cohort exceeded the annual incidence reported in the JNCR for most age brackets. Furthermore, the estimated prevalence rate in this study cohort (1.46%) exceeded that reported in all 17 large-scale studies, making the result an outlier (p = 0.005). These data indicate the possibility that the prevalence of CRC in PCa patients is greater than that in general average-risk adults, warranting further research in a prospective setting.

Development of an HDR-BT QA tool for source position verification.

PURPOSE: This study aimed to develop a high-dose-rate brachytherapy (HDR-BT) quality assurance (QA) tool for verification of source positions, and to report on its effectiveness. METHODS: We fabricated a cuboid phantom measuring 30 × 30×3 cm3 with spaces to embed Fletcher-Williamson tandem and ovoid applicators. Lead-based, cylindrically shaped radiopaque markers, which scatter radiation and blacken the Gafchromic® RTQA2 films placed on the applicators, were inserted into the phantom to determine the applicator tip and reference source positions. A three-dimensional image-guided brachytherapy (3D-IGBT) plan was generated, and the source positions on the film and radiation treatment planning system (RTPS) were verified with the tool. Source position errors were evaluated as the distance in the applicator axis direction between the source position and the center position of two radiopaque marker pairs. RESULTS: Source position errors on the film and RTPS were in good agreement with one another and were all within 0.5 mm for all applicators. Offset values of each applicator were in good agreement with the value determined in treatment planning (6 mm). The expanded measurement uncertainty of our QA tool was estimated to be 0.87 mm, with a coverage factor k of 2. CONCLUSIONS: Our new HDR-BT QA tool developed for comprehensive source position verification will be useful for cross checking actual source positions and planned source positions on the RTPS.

Stereotactic Body Radiotherapy for Pulmonary Oligometastasis from Colorectal Cancer.

BACKGROUND/AIM: A retrospective study was conducted to evaluate the safety and efficacy of stereotactic body radiotherapy (SBRT) for pulmonary oligometastasis from colorectal cancer (CRC). PATIENTS AND METHODS: Patients with pulmonary oligometastasis from CRC who were treated with SBRT between April 2010 and October 2018 were enrolled in this study. All patients underwent SBRT using Cyberknife® with a dose of 54-60 Gy in 3 fractions to 99% of the clinical target volume. The treatment efficacy was evaluated by the local control (LC) and overall survival (OS) rates. The toxicity was evaluated using the National Cancer Institute's Common Terminology Criteria for Adverse Events version 4.0. RESULTS: Twenty-six lesions in 20 patients were treated with SBRT. The median follow-up duration was 19 months (range=6-98 months). Local recurrence occurred in 6 of 26 lesions with a median follow-up of 12 months. The 2-year LC and OS rates were 65.8% and 88.6%, respectively. No patient developed ≥ grade 2 toxicity in the lung and other sites. CONCLUSION: Although very high doses were delivered to the tumors with SBRT, the LC of pulmonary metastasis from CRC was not satisfactory when compared to that for stage I primary non-small cell lung cancer reported in the literature.

Incidence and dose-volume relationship of radiation pneumonitis after concurrent chemoradiotherapy followed by durvalumab for locally advanced non-small cell lung cancer.

BACKGROUND AND PURPOSE: We investigated the incidence and dose-volume relationships of radiation pneumonitis (RP) after concurrent chemoradiotherapy (CCRT) followed by durvalumab for locally advanced non-small-cell lung cancer (LA-NSCLC). MATERIALS AND METHODS: We retrospectively analyzed records of 36 patients with LA-NSCLC who underwent CCRT followed by durvalumab. Incidence of RP was analyzed for correlations with clinical factors and dose-volume parameters of lung in radiotherapy. RESULTS: All patients received 60 Gy in 30 fractions of radiotherapy with concurrent chemotherapy. Over a median follow-up period of 7 months, incidence of grade ≥2 RP was 36% (including grade 3 RP: 5% and grade 5 RP: 3%). Age, sex, Brinkman index, and blood test results did not significantly differ between patients with grade ≥2 RP and grade ≤1 RP. Dose-volume parameters (lung volumes that received 5 Gy, 10 Gy, 20 Gy, 30 Gy, 40 Gy, 50 Gy, and mean lung dose) were significantly higher among patients with grade ≥2 RP compared with patients with grade ≤1 RP. CONCLUSION: Incidence of grade ≥2 RP was 36% after CCRT followed by durvalumab for LA-NSCLC, but did not significantly differ from those of patients treated with CCRT alone. Lung dose-volume parameters were significantly correlated with RP.

Pattern of Local Failure and its Risk Factors of Locally Advanced Non-small Cell Lung Cancer Treated With Concurrent Chemo-radiotherapy.

BACKGROUND/AIM: The treatment outcome of locally advanced non-small cell lung cancer (LA-NSCLC) has been improved over the past years but local failure is still common for these patients. The purpose of this study is to analyze the pattern of local failure and its risk factor of concurrent chemo-radiotherapy (CCRT) for locally advanced LA-NSCLC. PATIENTS AND METHODS: We evaluated 77 patients treated with CCRT for LA-NSCLC from July 2007 to December 2017 at our institution. Most of the patients were treated with 60 Gy in 30 fractions of radiotherapy and concurrent chemotherapy. The median follow-up time was 26 months. RESULTS: Among the 77 patients, 50 developed progressive disease during follow-up, including 14 with only local recurrence (LR), 10 with only distant metastasis and 26 with both. Of the 14 patients with only LR, 12 had primary tumor recurrence and 2 had recurrence in lymph nodes. A primary tumor volume of 50 cm3 was identified as the optimal cut-off value that was significantly correlated with primary tumor recurrence and overall survival. CONCLUSION: Primary tumor recurrence without lymph node and distant metastasis was observed in 12 patients (16%). Primary tumor volume of 50 cm3 was the optimal cut-off value for the prediction of primary tumor recurrence.

[A Case of Goblet Cell Carcinoid].

The patient was a 26-year-old female who had undergone conservative treatment for acute appendicitis at another clinic and was referred to our hospital for interval appendectomy. We performed a single-incision laparoscopic appendectomy, and the patient was diagnosed with goblet cell carcinoid(GCC)based on the postoperative pathological examination. Since GCC is considered a high-grade tumor, we performed a laparoscopic ileocolic resection with D3 lymphadenectomy. There was no residual disease or lymph node metastasis detected in the resected specimen. Patients with advanced GCC typically have poor prognosis, because GCC is characterized by peritoneal dissemination and lymph node metastasis. However, our findings suggested that an additional laparoscopic surgery could be one of the curative and safe treatment options for selected pa- tients with GCC.

[Spontaneous Disappearance of Undifferentiated Pleomorphic Primary Mesenteric Sarcoma-A Case Report].

The patient was a 49-year-old man with persistent fever since the introduction of hemodialysis(HD). Vomiting and abdominal swelling appeared 4 months after initiating hemodialysis. Computed tomography(CT)scan revealed a tumor measuring 9 cm, and disorders of passage from the jejunum. Surgery was performed, and resection was impossible because of peritoneal dissemination. Histopathological examination of the disseminated nodes suggested an undifferentiated pleomorphic sarcoma. Postoperatively, drainage from the gastric fistula was approximately 2,000mL/day. Chemotherapy was considered impossible because of HD, and palliative therapy was selected. However, the volume of drainage from the gastric fistula gradually decreased, and the disorders of passage reduced. CT scan confirmed marked reduction in the size of the intraperitoneal tumor and its subsequent disappearance. At the 2-year-and-5-month postoperative follow-up, no relapses were observed, and the course had been uneventful. Undifferentiated pleomorphic sarcomas develop in the soft tissue of adults and have a poor prognosis. However, mesenteric development is rare. Total tumorectomy is the first choice of treatment. A consensus on the usefulness of chemotherapy or radiotherapy has not been reached. Furthermore, no studies have reported spontaneous tumor disappearance in the absence of treatment. Here, we report a case of minor undifferentiated primary mesenteric sarcoma and its spontaneous disappearance and review the literature.

Laparoscopic distal gastrectomy in a patient with gastric cancer complicated by eventration of the diaphragm: A case report.

Diaphragmatic eventration refers to an abnormal elevation of the diaphragm. Here, we report the case of a patient with gastric cancer who underwent successful laparoscopic distal gastrectomy despite the presence of diaphragmatic eventration. The patient was a 72-year-old man diagnosed with early gastric cancer in the antrum, as detected by upper gastrointestinal endoscopy. Preoperative imaging revealed an elevation of the left side of the diaphragm, which was diagnosed as diaphragmatic eventration. Laparoscopic surgery is beneficial for obtaining an optimal field of view. However, there are critical points that must be considered when laparoscopic distal gastrectomy is performed in patients with gastric cancer complicated by diaphragmatic eventration. There were difficulties that affected manipulation because the elevated diaphragm drew the intraperitoneal organs into the thoracic cavity, causing displacement of the normal anatomical position. We found it beneficial to secure the lesser curvature of the stomach given the possible effects of gastric deformation.