Development of deep learning framework for anatomical landmark detection and guided dissection line during laparoscopic cholecystectomy.

Background: Bile duct injuries during laparoscopic cholecystectomy can arise from misinterpretation of biliary anatomy, leading to dissection in improper areas. The integration of a deep learning framework into laparoscopic procedures offers the potential for real-time anatomical landmark recognition, ensuring accurate dissection. The objective of this study is to develop a deep learning framework that can precisely identify anatomical landmarks, including Rouviere's sulcus and the liver base of segment IV, and provide a guided dissection line during laparoscopic cholecystectomy. Methods: We retrospectively collected 40 laparoscopic cholecystectomy videos and extracted 80 images form each video to establish the dataset. Three surgeons annotated the bounding boxes of anatomical landmarks on a total of 3200 images. The YOLOv7 model was trained to detect Rouviere's sulcus and the liver base of segment IV as anatomical landmarks. Additionally, the guided dissection line was generated between these two landmarks by the proposed algorithm. To evaluate the performance of the detection model, mean average precision (mAP), precision, and recall were calculated. Furthermore, the accuracy of the guided dissection line was evaluated by three surgeons. The performance of the detection model was compared to the scaled-YOLOv4 and YOLOv5 models. Finally, the proposed framework was deployed in the operating room for real-time detection and visualization. Results: The overall performance of the YOLOv7 model on validation set and testing set were 98.1 % and 91.3 %, respectively. Surgeons accepted the visualization of guide dissection line with a rate of 95.71 %. In the operating room, the well-trained model accurately identified the anatomical landmarks and generated the guided dissection line in real-time. Conclusions: The proposed framework effectively identifies anatomical landmarks and generates a guided dissection line in real-time during laparoscopic cholecystectomy. This research underscores the potential of using deep learning models as computer-assisted tools in surgery, providing an assistant tool to accommodate with surgeons.

Propensity score-matched comparison of safety outcomes between high-risk and low-risk patients towards early hospital discharge after laparoscopic cholecystectomy.

Background: Laparoscopic cholecystectomy (LC), a common treatment for symptomatic gallstones, has demonstrated safety in low-risk patients. However, existing data are scarce regarding the safety of LC in high-risk patients and the feasibility of early hospital discharge. Materials and methods: This retrospective study included 2296 patients diagnosed with symptomatic gallstones who underwent LC at a tertiary care centre from January 2009 through December 2019. The authors employed propensity score matching to mitigate bias between groups. Statistical significance was set at P less than 0.05. Results: The median age of the patients was 56 years (range 46-67), with a mean BMI of 25.2±4.3 kg/m2. Patients were classified as: American Society of Anesthesiologists (ASA) I (19.7%), II (68.3%), III (12.0%), and IV (0%). ASA I-II included low surgical risk patients (88%) and ASA III-IV comprised high-risk patients (12%). The LC-related 30-day reoperative rate was 0.2% and the readmission rate was 0.87%. Nine patients (0.4%) sustained major bile duct injuries, resulting in a conversion rate of 2.4%. The postoperative mortality rate was 0.04%, and the mean hospitalization time was 3.5 days. Patients in the high-risk group with a history of acute cholecystitis exhibited greater estimated blood loss, longer operative times, and were significantly more likely to be in the longer-stay group. Conclusion: These findings suggest that LC can be conducted safely on high-risk patients, and early hospital discharge is achievable. However, specific factors, such as a history of acute cholecystitis, may result in prolonged hospitalization owing to increased blood loss and longer operative times.

Development of a deep learning model for safe direct optical trocar insertion in minimally invasive surgery: an innovative method to prevent trocar injuries.

BACKGROUND: Direct optical trocar insertion is a common procedure in laparoscopic minimally invasive surgery. However, misinterpretations of the abdominal wall anatomy can lead to severe complications. Artificial intelligence has shown promise in surgical endoscopy, particularly in the employment of deep learning models for anatomical landmark identification. This study aimed to integrate a deep learning model with an alarm system algorithm for the precise detection of abdominal wall layers during trocar placement. METHOD: Annotated bounding boxes and assigned classes were based on the six layers of the abdominal wall: subcutaneous, anterior rectus sheath, rectus muscle, posterior rectus sheath, peritoneum, and abdominal cavity. The cutting-edge YOLOv8 model was combined with a deep learning detector to train the dataset. The model was trained on still images and inferenced on laparoscopic videos to ensure real-time detection in the operating room. The alarm system was activated upon recognizing the peritoneum and abdominal cavity layers. We assessed the model's performance using mean average precision (mAP), precision, and recall metrics. RESULTS: A total of 3600 images were captured from 89 laparoscopic video cases. The proposed model was trained on 3000 images, validated with a set of 200 images, and tested on a separate set of 400 images. The results from the test set were 95.8% mAP, 89.8% precision, and 91.7% recall. The alarm system was validated and accepted by experienced surgeons at our institute. CONCLUSION: We demonstrated that deep learning has the potential to assist surgeons during direct optical trocar insertion. During trocar insertion, the proposed model promptly detects precise landmark references in real-time. The integration of this model with the alarm system enables timely reminders for surgeons to tilt the scope accordingly. Consequently, the implementation of the framework provides the potential to mitigate complications associated with direct optical trocar placement, thereby enhancing surgical safety and outcomes.

Laparoscopic modified fundus-down cholecystectomy technique: an alternative method for performing a safe laparoscopic cholecystectomy: how to article.

Modified fundus-down cholecystectomy is a surgical procedure used to treat patients diagnosed with benign gallbladder disease. This technique begins with Calot's triangle dissection and attempts to identify key structures such as the cystic artery and duct. Subsequently, fundus-down dissection is performed to separate the gallbladder from the cystic plate. The cystic artery and duct are the final structures that are clipped and cut. In this study, the authors discuss the success and complication rates of this treatment based on their 10-year experience at a tertiary hospital in southern Thailand. Objectives: This study aimed to compare the operative outcomes of conventional laparoscopic cholecystectomy (LC) and modified fundus-down techniques regarding postoperative complications and consequences. Methods: A retrospective analysis of single-centre data from 2010 to 2022 was conducted at our hospital. All patients with gallstone disease who underwent conventional LC or modified fundus-down cholecystectomy were included in the study. The primary outcomes of this study were the incidence of major bile duct injury and the need for further intervention or surgical correction. Results: From a total of 1993 patients who were surveyed, 1612 patients underwent conventional LC and 381 underwent laparoscopic modified fundus-down cholecystectomy. In terms of conversion rate, estimated blood loss, length of hospital stay, and complication rate, there were no differences between the conventional LC and the modified fundus-down approach. However, modified fundus-down cholecystectomy reduced the operative time. The authors collected data from each patient's sign-in to extubation time (P<0.001). The postoperative complications (P=0.120) and conversion rates (P=0.904) were similar. Conclusion: Laparoscopic modified fundus-down cholecystectomy can be performed in simple and complex cases, including cases of severe fibrosis of the hepatocystic triangle. The study showed that this alternative technique could reduce operative time compared to the conventional technique with no difference in complications, especially common bile duct injury, postoperative common bile duct stones, and postoperative pancreatitis.

A Clinical Prediction Model for Breast Cancer in Women Having Their First Mammogram.

BACKGROUND: Digital mammography is the most efficient screening and diagnostic modality for breast cancer (BC). However, the technology is not widely available in rural areas. This study aimed to construct a prediction model for BC in women scheduled for their first mammography at a breast center to prioritize patients on waiting lists. METHODS: This retrospective cohort study analyzed breast clinic data from January 2013 to December 2017. Clinical parameters that were significantly associated with a BC diagnosis were used to construct predictive models using stepwise multiple logistic regression. The models' discriminative capabilities were compared using receiver operating characteristic curves (AUCs). RESULTS: Data from 822 women were selected for analysis using an inverse probability weighting method. Significant risk factors were age, body mass index (BMI), family history of BC, and indicated symptoms (mass and/or nipple discharge). When these factors were used to construct a model, the model performance according to the Akaike criterion was 1387.9, and the AUC was 0.82 (95% confidence interval: 0.76-0.87). CONCLUSION: In a resource-limited setting, the priority for a first mammogram should be patients with mass and/or nipple discharge, asymptomatic patients who are older or have high BMI, and women with a family history of BC.

AI-Assisted Assessment of Wound Tissue with Automatic Color and Measurement Calibration on Images Taken with a Smartphone.

Wound assessment is essential for evaluating wound healing. One cornerstone of wound care practice is the use of clinical guidelines that mandate regular documentation, including wound size and wound tissue composition, to determine the rate of wound healing. The traditional method requires wound care professionals to manually measure the wound area and tissue composition, which is time-consuming, costly, and difficult to reproduce. In this work, we propose an approach for automatic wound assessment that incorporates automatic color and measurement calibration and artificial intelligence algorithms. Our approach enables the comparison of images taken at different times, even if they were taken under different lighting conditions, distances, lenses, and camera sensors. We designed a calibration chart and developed automatic algorithms for color and measurement calibration. The wound area and wound composition on the images were annotated by three physicians with more than ten years of experience. Deep learning models were then developed to mimic what the physicians did on the images. We examined two network variants, U-Net with EfficientNet and U-Net with MobileNetV2, on wound images with a size of 1024 × 1024 pixels. Our best-performing algorithm achieved a mean intersection over union (IoU) of 0.6964, 0.3957, 0.6421, and 0.1552 for segmenting a wound area, epithelialization area, granulation tissue, and necrotic tissue, respectively. Our approach was able to accurately segment the wound area and granulation tissue but was inconsistent with respect to the epithelialization area and necrotic tissue. The calibration chart, which helps calibrate colors and scales, improved the performance of the algorithm. The approach could provide a thorough assessment of the wound, which could help clinicians tailor treatment to the patient's condition.

Impact of preoperative esophagogastroduodenoscopy in patients undergoing bariatric surgery and development of a model to predict clinically significant abnormal endoscopic findings.

BACKGROUND: Preoperative esophagogastroduodenoscopy (EGD) in patients undergoing bariatric surgery can help surgeons detect abnormalities in the upper gastrointestinal (UGI) tract that may require a change in surgical plan. However, the need for EGD before bariatric surgery is controversial. OBJECTIVES: We aimed to determine the prevalence of UGI abnormalities and evaluate the predictive factors of abnormal findings that require a change in surgical plan or cause a delay in surgical treatment in patients undergoing bariatric surgery and develop a prediction model. METHODS: The medical records from January 2012 to July 2020 were retrospectively reviewed in patients who underwent EGD before bariatric surgery. The EGD findings were classified into four groups based on their effects on management. Group 1 had normal findings. Group 2 had abnormal findings that did not require a change in surgical management. Group 3 had abnormal findings that required a change in the surgical plan or caused a delay in surgical treatment. Group 4 had contraindications to surgery. Predictive factors for Groups 3 and 4 were analyzed using univariate and multivariate analyses. A model visualized as a nomogram was developed based on significant factors. Discrimination and calibration were evaluated. RESULTS: A total of 461 patient records (63.8% female) were reviewed. The mean age was 35.1 ± 11.2 years and the mean BMI was 47.7 ± 8.7 kg/m2. The prevalence of endoscopic findings in Groups 1, 2, 3, and 4 were 42.5%, 35.6%, 21.9%, and 0%, respectively. The most common abnormal findings were non-erosive gastritis (31.2%) followed by Helicobacter pylori infection (18.7%) and hiatal hernia (10.2%). Male sex and NSAID use were significantly associated with detection of lesions in Group 3 either on univariate or multivariate analysis, while type 2 diabetes mellitus (T2DM) was a significant protective factor on multivariate analysis. On subgroup analysis in patients ≥ 40 years old, multivariate analysis revealed age, BMI, and NSAID use were significantly associated with the detection of lesions in Group 3, while T2DM was still a significant protective factor. A nomogram to predict lesions in Group 3 for this subgroup was developed and showed good discrimination (C-statistics 0.737, 95% CI 0.721‒0.752). CONCLUSION: A high prevalence of abnormal endoscopic findings was observed in Thai patients who are undergoing bariatric surgery. Preoperative EGD screening is helpful in detecting UGI abnormalities that require a change in the surgical decision plan. The new nomogram may help rational utilization of EGD prior to bariatric surgery.

Laparoscopic trans-hiatal approach in a rare case of double giant epiphrenic diverticula: A case report.

The transthoracic approach for esophageal diverticulectomy saves dissection but limits the amount of antireflux surgery that can be done. However, the laparoscopic trans-hiatal approach usually limits access to large diverticula located at the mid-thoracic level. We describe a 67-year-old Thai man who had two giant epiphrenic diverticula (ED) (11 × 7.6 cm and 3.2 × 3.4 cm). The highest part of the fundus of the larger ED was 12.6 cm above the hiatus. Preoperative esophageal manometry and an upper gastrointestinal study showed no obvious features of motility disorder. In addition, the clinical evaluation of the esophagogastroduodenoscopy revealed a hiatal hernia. Laparoscopic trans-hiatal esophageal diverticulectomy with intraoperative assessment by endoscopy was carried out. A partial wrap using Dor fundoplication was constructed without myotomy. There were no postoperative related complications. We believe that tailor-made operations in unusual cases can be safe using combined techniques to reduce postoperative morbidity.

Propensity Score-Matched Analysis of Laparoscopic versus Open Surgery for Non-Metastatic Rectal Cancer.

BACKGROUND: Rectal cancer is a pervasive type of malignancy that accounts for one-third of colorectal cancers worldwide. Several studies have assessed the use of laparoscopic surgery as a treatment option. However, there is an ongoing debate regarding its oncological safety. METHODS: This retrospective study included 270 patients with non-metastatic rectal cancer who underwent either laparoscopic resection (LR, n = 93) or open resection (OR, n = 177) in an academic medical center. The primary outcomes were overall survival (OS) and disease-free survival (DFS), whereas the secondary outcome was postoperative complications. We performed propensity score analyses and compared outcomes. Univariate survival analyses using Kaplan-Meier plots and Cox proportional hazard regression models were also conducted. RESULTS: In the propensity score matching analyses, 93 LR- and 93 OR-matched patients were compared. The overall median follow-up time was 3.95 years (range, 1.98‒5.55 years). The 3-year OS was similar between the groups (LR 79.1% vs OR 79.2%, p = 0.82). Meanwhile, the DFS rate was also comparable between the groups (LR 77.8% vs OR 73.2%, p = 0.53). No significant differences in operative blood loss or hospital stay between the groups were observed (150 vs 150 mL, p = 0.74; 9 vs 10 days, p = 0.077, respectively). Also, no difference was found in postoperative complications between the groups (p = 0.23). However, LR was associated with a longer operative time than OR (455 vs 356 min, p < 0.001) and the number of lymph nodes harvested in LR was slightly fewer than OR (10 vs 11, p = 0.045). CONCLUSION: LR of rectal cancer is safe, feasible, and comparable to standard OR in terms of the oncologic outcomes. However, LR required longer operative times. A well-designed prospective study with a large number of participants and long follow-up period is needed to show significant differences between the two groups.
.

Prediction for Breast Cancer in BI-RADS Category 4 Lesion Categorized by Age and Breast Composition of Women in Songklanagarind Hospital.

BACKGROUND: Older age and dense breast are the important risk factors for breast cancer. The ACR BI-RADS lexicon 5th edition does not mention how patient age and breast density may affect the category assessment. The aim of this study was to investigate whether patient age and breast density influence the positive predictive value (PPV) of mammographic and ultrasonographic findings categorized as BI-RADS category 4 and subcategories 4a, 4b, and 4c among female patients. MATERIALS AND METHODS: A retrospective study was conducted in Songklanagarind Hospital between January 1, 2016 and December 31, 2017 in female patients older than 18 years who had breast lesions categorized as BI-RADS category 4 and subcategories 4a, 4b, 4c. A total of 961 breast lesions consisted of 772 (80.33%) benign lesions and 189 (19.67%) malignant lesions. Categorization was done in each lesion based on age ranges of ≤35 years, >35 to 60 years, and >60 years and breast density according to mammographic breast composition. The PPV for each BI-RADS category was calculated based on the pathological diagnoses and were compared using the chi-square test. RESULTS: The overall PPV in each subcategory was in the reference range. The PPV increased with increasing age: 4% vs. 22.63% vs. 36.67% for category 4 (p-value=0.01); 0% vs. 5.81% vs. 6.88% for subcategory 4a (p-value=0.002); 6.67% vs. 26.62% vs. 51.35% for subcategory 4b (p-value=0.001); and 33.33% vs. 76.92% vs. 81.82% for subcategory 4c (p-value=0.02). An association was not found between PPV and breast density. CONCLUSION: A significantly positive association was found between PPV and age in patients in BI-RADS subcategories 4a, 4b, and 4c. This study could not determine that mammographic breast composition according to the ACR BI-RADS 5th edition was associated with PPV due to improper sample distribution.

Mimicked cartilage scaffolds of silk fibroin/hyaluronic acid with stem cells for osteoarthritis surgery: Morphological, mechanical, and physical clues.

Osteoarthritis is a critical disease that comes from degeneration of cartilage tissue. In severe cases surgery is generally required. Tissue engineering using scaffolds with stem cell transplantation is an attractive approach and a challenge for orthopedic surgery. For sample preparation, silk fibroin (SF)/hyaluronic acid (HA) scaffolds in different ratios of SF/HA (w/w) (i.e., 100:0, 90:10, 80:20, and 70:30) were formed by freeze-drying. The morphological, mechanical, and physical clues were considered in this research. The morphological structure of the scaffolds was observed by scanning electron microscope. The mechanical and physical properties of the scaffolds were analyzed by compressive and swelling ratio testing, respectively. For the cell experiments, scaffolds were seeded and cultured with human umbilical cord-derived mesenchymal stem cells (HUMSCs). The cultured scaffolds were tested for cell viability, histochemistry, immunohistochemistry, and gene expression. The SF with HA scaffolds showed regular porous structures. Those scaffolds had a soft and elastic characteristic with a high swelling ratio and water uptake. The SF/HA scaffolds showed a spheroid structure of the cells in the porous structure particularly in the SF80 and SF70 scaffolds. Cells could express Col2a, Agg, and Sox9 which are markers for chondrogenesis. It could be deduced that SF/HA scaffolds showed significant clues for suitability in cartilage tissue engineering and in surgery for osteoarthritis.