T2-weighted imaging-based radiomic-clinical machine learning model for predicting the differentiation of colorectal adenocarcinoma

BACKGROUND The study on predicting the differentiation grade of colorectal cancer(CRC)based on magnetic resonance imaging(MRI)has not been reported yet.Developing a non-invasive model to predict the differentiation grade of CRC is of great value.AIM To develop and validate machine learning-based models for predicting the differ-entiation grade of CRC based on T2-weighted images(T2WI).METHODS We retrospectively collected the preoperative imaging and clinical data of 315 patients with CRC who underwent surgery from March 2018 to July 2023.Patients were randomly assigned to a training cohort(n=220)or a validation cohort(n=95)at a 7:3 ratio.Lesions were delineated layer by layer on high-resolution T2WI.Least absolute shrinkage and selection operator regression was applied to screen for radiomic features.Radiomics and clinical models were constructed using the multilayer perceptron(MLP)algorithm.These radiomic features and clinically relevant variables(selected based on a significance level of P<0.05 in the training set)were used to construct radiomics-clinical models.The performance of the three models(clinical,radiomic,and radiomic-clinical model)were evaluated using the area under the curve(AUC),calibration curve and decision curve analysis(DCA).RESULTS After feature selection,eight radiomic features were retained from the initial 1781 features to construct the radiomic model.Eight different classifiers,including logistic regression,support vector machine,k-nearest neighbours,random forest,extreme trees,extreme gradient boosting,light gradient boosting machine,and MLP,were used to construct the model,with MLP demonstrating the best diagnostic performance.The AUC of the radiomic-clinical model was 0.862(95%CI:0.796-0.927)in the training cohort and 0.761(95%CI:0.635-0.887)in the validation cohort.The AUC for the radiomic model was 0.796(95%CI:0.723-0.869)in the training cohort and 0.735(95%CI:0.604-0.866)in the validation cohort.The clinical model achieved an AUC of 0.751(95%CI:0.661-0.842)in the training cohort and 0.676(95%CI:0.525-0.827)in the validation cohort.All three models demonstrated good accuracy.In the training cohort,the AUC of the radiomic-clinical model was significantly greater than that of the clinical model(P=0.005)and the radiomic model(P=0.016).DCA confirmed the clinical practicality of incorporating radiomic features into the diagnostic process.CONCLUSION In this study,we successfully developed and validated a T2WI-based machine learning model as an auxiliary tool for the preoperative differentiation between well/moderately and poorly differentiated CRC.This novel approach may assist clinicians in personalizing treatment strategies for patients and improving treatment efficacy.

Laparoscopic resection and endoscopic submucosal dissection for treating gastric ectopic pancreas

BACKGROUND Gastric ectopic pancreas(GEP)is a rare developmental abnormality that refers to the existence of pancreatic tissue in the stomach with no anatomical relationship with the main pancreas.It is usually difficult to diagnose through histological examination,and the choice of treatment method is crucial.AIM To describe the endoscopic ultrasound characteristics of GEP and evaluate the value of laparoscopic resection(LR)and endoscopic submucosal dissection(ESD).METHODS Forty-nine patients with GEP who underwent ESD and LR in the Second Affiliated Hospital of Fujian Medical University from May 2018 to July 2023 were retrospectively included.Data on clinical characteristics,endoscopic ultrasonography(EUS),ESD,and LR were collected and analyzed.The characteristics of EUS and the efficacy of the two treatments were analyzed.RESULTS The average age of the patients was 43.31±13.50 years,and the average maximum diameter of the lesions was 1.55±0.70 cm.The lesion originated from the mucosa in one patient(2.04%),from the submucosa in 42 patients(85.71%),and from the muscularis propria in 6 patients(12.25%).Twenty-nine patients(59.20%)with GEP showed umbilical depression on endoscopy.The most common initial symptom of GEP was abdominal pain(40.82%).Tumor markers,including carcinoembryonic antigen(CEA)and carbohydrate antigen 19-9(CA19-9),were generally within the normal range.One patient(2.04%)with GEP had increased CEA and CA-19-9 levels.However,no cancer tissue was found on postoperative pathological examination,and tumor markers returned to normal levels after resecting the lesion.There was no significant difference in surgery duration(72.42±23.84 vs 74.17±12.81 min)or hospital stay(3.70±0.91 vs 3.83±0.75 d)between the two methods.LR was more often used for patients with larger tumors and deeper origins.The amount of bleeding was significantly higher in LR than in ESD(11.28±16.87 vs 16.67±8.76 mL,P<0.05).Surgery was associated with complete resection of the lesion without any serious complications;there were no cases of recurrence during the follow-up period.CONCLUSION GEP has unique characteristics in EUS.LR and ESD seem to be good choices for treating GEP.LR is better for large GEP with a deep origin.However,due to the rarity of GEP,multicenter large-scale studies are needed to describe its characteristics and evaluate the safety of LR and ESD.

Development and validation of a nomogram for preoperative prediction of tumor deposits in colorectal cancer

BACKGROUND Based on the clinical data of colorectal cancer(CRC)patients who underwent surgery at our institution,a model for predicting the formation of tumor deposits(TDs)in this patient population was established.AIM To establish an effective model for predicting TD formation,thus enabling clinicians to identify CRC patients at high risk for TDs and implement personalized treatment strategies.METHODS CRC patients(n=645)who met the inclusion criteria were randomly divided into training(n=452)and validation(n=193)cohorts using a 7:3 ratio in this retrospective analysis.Least absolute shrinkage and selection operator regression was employed to screen potential risk factors,and multivariable logistic regression analysis was used to identify independent risk factors.Subsequently,a predictive model for TD formation in CRC patients was constructed based on the independent risk factors.The discrimination ability of the model,its consistency with actual results,and its clinical applicability were evaluated using receiveroperating characteristic curves,area under the curve(AUC),calibration curves,and decision curve analysis(DCA).RESULTS Thirty-four(7.5%)patients with TDs were identified in the training cohort based on postoperative pathological specimens.Multivariate logistic regression analysis identified female sex,preoperative intestinal obstruction,left-sided CRC,and lymph node metastasis as independent risk factors for TD formation.The AUCs of the nomogram models constructed using these variables were 0.839 and 0.853 in the training and validation cohorts,respectively.The calibration curve demonstrated good consistency,and the training cohort DCA yielded a threshold probability of 7%-78%.CONCLUSION This study developed and validated a nomogram with good predictive performance for identifying TDs in CRC patients.Our predictive model can assist surgeons in making optimal treatment decisions.

Splenosis masquerading as gastric stromal tumor: A case report

BACKGROUND Splenosis is a rare benign disease that often disguises itself as a malignant tumor.There are few articles providing a comprehensive description of splenosis,especially cases located in the stomach being treated by laparoscopic surgery.CASE SUMMARY A 44-year-old man presented with recurrent upper abdominal pain for more than half a year.The patient had splenic rupture caused by trauma more than 10 years ago and underwent splenectomy.An abdominal contrast-enhanced computed tomography scan revealed an irregular soft tissue density.Gastroscopy revealed an approximately 3.0 cm×3.0 cm mucosal eminence at the posterior wall of the upper segment of the gastric body.Biopsy was not performed since the lesion was found under the mucosa and the gastric mucosa appeared normal.According to these findings,a diagnosis of gastric stromal tumor was made,although a definitive differential diagnosis was not known before surgery.When laparoscopic resection of the gastric stromal tumor was performed,an astonishing finding was made when postoperative pathology showed that the lesion comprised typical spleen tissue.CONCLUSION This case highlights the strong similarities between splenosis and malignant tumors.A detailed medical history combined with various effective auxiliary examinations can help improve differential diagnosis.

Analysis of 20 patients with laparoscopic extended right colectomy

BACKGROUND Currently,the standard surgical procedure for right colon cancer is complete mesocolic excision.Whether preventive extended lymph node dissection for colon cancer located in the hepatic flexure or right transverse colon should be performed remains controversial because the safety and effectiveness of the operation have not been proven,and infrapyloric lymph nodes(No.206)and lymph nodes in the greater curvature of the stomach(No.204)have not been strictly defined and distinguished as surgical indicators in previous studies.AIM To analyze the metastatic status of infrapyloric lymph nodes and lymph nodes of the greater curvature of the stomach and perioperative complications and systematically evaluate the feasibility and safety of laparoscopic extended right colectomy using prospective data collected retrospectively.METHODS The study was a clinical study.Twenty patients with colon cancer who underwent laparoscopic extended right colon resection in our hospital from June 2020 to May 2021 were included.RESULTS Among the patients who underwent extended right colon resection,there were no intraoperative complications or conversion to laparotomy;2 patients had gastrocolic ligament lymph node metastasis,and 5 patients had postoperative complications.The patients with postoperative complications received conservative treatment.CONCLUSION Laparoscopic extended right colon resection is safe.However,malignant tumors located in the liver flexure or the right-side transverse colon are more likely to metastasize to the gastrocolic ligament lymph nodes,and notably,the incidence of gastroparesis was high.The number of patients was small,and the follow-up time was short.It is necessary to further increase the sample size to evaluate the No.204 and No.206 lymph node metastasis rates and the long-term survival impact.

Nomogram for predicting chylous ascites after right colectomy

BACKGROUND Chylous ascites following right colectomy has a high incidence which is a critical challenge.At present,there are few studies on the factors affecting chylous ascites after right colectomy and especially after D3 Lymphadenectomy.A predictive model for chylous ascites has not yet been established.Therefore,we created the first nomogram to predict the incidence of chylous ascites after right hemicolectomy.AIM To analyze the risk factors for chylous ascites after right colectomy and establish a nomogram to predict the incidence of chylous ascites.METHODS We retrospectively collected patients who underwent right hemicolectomy between January 2012 and May 2021 and were pathologically diagnosed with cancer.Multivariate logistic regression was used to analyze the influencing factors of chylous ascites and a nomogram was established.The predictive ability was assessed by the area under the receiver operating characteristic(ROC)curve.RESULTS Operative time,the type of operation(standard or extended),the number of lymph nodes retrieved,and somatostatin administration were considered important risk factors.Multivariate logistic regression and nomograms can be used to accurately predict whether chylous ascites occurs.The area under the ROC curve of the model is 0.770.The C-statistic of this model is 0.770 which indicates that it has a relatively moderate ability to predict the risk of chylous ascites.CONCLUSION We found a novel set of risk factors,created a nomogram,and validated it.The nomogram had a relatively accurate forecasting ability for chylous ascites after right hemicolectomy and can be used as a reference for risk assessment of chylous ascites and whether to prevent it after surgery.