Development of an image-based Random Forest classifier for prediction of surgery duration of laparoscopic sigmoid resections.

PURPOSE: Sigmoid diverticulitis is a disease with a high socioeconomic burden, accounting for a high number of left-sided colonic resections worldwide. Modern surgical scheduling relies on accurate prediction of operation times to enhance patient care and optimize healthcare resources. This study aims to develop a predictive model for surgery duration in laparoscopic sigmoid resections, based on preoperative CT biometric and demographic patient data. METHODS: This retrospective single-center cohort study included 85 patients who underwent laparoscopic sigmoid resection for diverticular disease. Potentially relevant procedure-specific anatomical parameters recommended by a surgical expert were measured in preoperative CT imaging. After random split into training and test set (75% / 25%) multiclass logistic regression was performed and a Random Forest classifier was trained on CT imaging parameters, patient age, and sex in the training cohort to predict categorized surgery duration. The models were evaluated in the test cohort using established performance metrics including receiver operating characteristics area under the curve (AUROC). RESULTS: The Random Forest model achieved a good average AUROC of 0.78. It allowed a very good prediction of long (AUROC = 0.89; specificity 0.71; sensitivity 1.0) and short (AUROC = 0.81; specificity 0.77; sensitivity 0.56) procedures. It clearly outperformed the multiclass logistic regression model (AUROC: average = 0.33; short = 0.31; long = 0.22). CONCLUSION: A Random Forest classifier trained on demographic and CT imaging biometric patient data could predict procedure duration outliers of laparoscopic sigmoid resections. Pending validation in a multicenter study, this approach could potentially improve procedure scheduling in visceral surgery and be scaled to other procedures.

Associations of body composition parameters with postoperative outcome and perineural tumour invasion after oncological pancreatic resection.

BACKGROUND: Pancreatic cancer is often accompanied by wasting conditions. While surgery is the primary curative approach, it poses a substantial risk of postoperative complications, hindering subsequent treatments. Therefore, identifying patients at high risk for complications and optimizing their perioperative general condition is crucial. Sarcopenia and other body composition abnormalities have shown to adversely affect surgical and oncological outcomes in various cancer patients. As most pancreatic tumours are located close to the neuronal control centre for the digestive tract, it is possible that neural infiltration in this area deranges bowel functions and contributes to malabsorption and malnutrition and ultimately worsen sarcopenia and weight loss. METHODS: A retrospective analysis of CT scans was performed for pancreatic cancer patients who underwent surgical tumour resection at a single high-volume centre from 2007 to 2023. Sarcopenia prevalence was assessed by skeletal muscle index (SMI), and visceral obesity was determined by the visceral adipose tissue area (VAT). Obesity and malnutrition were determined by the GLIM criteria. Sarcopenic obesity was defined as simultaneous sarcopenia and obesity. Postoperative complications, mortality and perineural tumour invasion, were compared among patients with body composition abnormalities. RESULTS: Of 437 patients studied, 46% were female, the median age was 69 (61;74) years. CT analysis revealed 54.9% of patients with sarcopenia, 23.7% with sarcopenic obesity and 45.9% with visceral obesity. Sarcopenia and sarcopenic obesity were more prevalent in elderly and male patients. Postoperative surgical complications occurred in 67.7% of patients, most of which were mild (41.6%). Severe complications occurred in 22.7% of cases and the mortality rate was 3.4%. Severe postoperative complications were significantly more common in patients with sarcopenia or sarcopenic obesity. Visceral obesity or malnutrition based on BMI alone, did not significantly impact complications. Perineural invasion was found in 80.1% of patients and was unrelated to malnutrition or body composition parameters. CONCLUSIONS: This is the first and largest study evaluating the associations of CT-based body mass analysis with surgical outcome and histopathological perineural tumour invasion in pancreatic cancer patients. The results suggest that elderly and male patients are at high risk for sarcopenia and should be routinely evaluated by CT before undergoing pancreatic surgery, irrespective of their BMI. Confirmation of the results in prospective studies is needed to assess if pancreatic cancer patients with radiographic sarcopenia benefit from preoperative amelioration of muscle mass and function by exercise and nutritional interventions.

Evaluation of GPT-4's Chest X-Ray Impression Generation: A Reader Study on Performance and Perception.

Exploring the generative capabilities of the multimodal GPT-4, our study uncovered significant differences between radiological assessments and automatic evaluation metrics for chest x-ray impression generation and revealed radiological bias.

Fractal analysis of extracellular matrix for observer-independent quantification of intestinal fibrosis in Crohn's disease.

Prevention of intestinal fibrosis remains an unresolved problem in the treatment of Crohn's disease (CD), as specific antifibrotic therapies are not yet available. Appropriate analysis of fibrosis severity is essential for assessing the therapeutic efficacy of potential antifibrotic drugs. The aim of this study was to develop an observer-independent method to quantify intestinal fibrosis in surgical specimens from patients with CD using structural analysis of the extracellular matrix (ECM). We performed fractal analysis in fibrotic and control histological sections of patients with surgery for CD (n = 28). To specifically assess the structure of the collagen matrix, polarized light microscopy was used. A score to quantify collagen fiber alignment and the color of the polarized light was established. Fractal dimension as a measure for the structural complexity correlated significantly with the histological fibrosis score whereas lacunarity as a measure for the compactness of the ECM showed a negative correlation. Polarized light microscopy to visualize the collagen network underlined the structural changes in the ECM network in advanced fibrosis. In conclusion, observer-independent quantification of the structural complexity of the ECM by fractal analysis is a suitable method to quantify the degree of intestinal fibrosis in histological samples from patients with CD.