Predicting rectal cancer tumor budding grading based on MRI and CT with multimodal deep transfer learning: A dual-center study.

Objective: To investigate the effectiveness of a multimodal deep learning model in predicting tumor budding (TB) grading in rectal cancer (RC) patients. Materials and methods: A retrospective analysis was conducted on 355 patients with rectal adenocarcinoma from two different hospitals. Among them, 289 patients from our institution were randomly divided into an internal training cohort (n = 202) and an internal validation cohort (n = 87) in a 7:3 ratio, while an additional 66 patients from another hospital constituted an external validation cohort. Various deep learning models were constructed and compared for their performance using T1CE and CT-enhanced images, and the optimal models were selected for the creation of a multimodal fusion model. Based on single and multiple factor logistic regression, clinical N staging and fecal occult blood were identified as independent risk factors and used to construct the clinical model. A decision-level fusion was employed to integrate these two models to create an ensemble model. The predictive performance of each model was evaluated using the area under the curve (AUC), DeLong's test, calibration curve, and decision curve analysis (DCA). Model visualization Gradient-weighted Class Activation Mapping (Grad-CAM) was performed for model interpretation. Results: The multimodal fusion model demonstrated superior performance compared to single-modal models, with AUC values of 0.869 (95% CI: 0.761-0.976) for the internal validation cohort and 0.848 (95% CI: 0.721-0.975) for the external validation cohort. N-stage and fecal occult blood were identified as clinically independent risk factors through single and multivariable logistic regression analysis. The final ensemble model exhibited the best performance, with AUC values of 0.898 (95% CI: 0.820-0.975) for the internal validation cohort and 0.868 (95% CI: 0.768-0.968) for the external validation cohort. Conclusion: Multimodal deep learning models can effectively and non-invasively provide individualized predictions for TB grading in RC patients, offering valuable guidance for treatment selection and prognosis assessment.

Multi-modality radiomics model predicts axillary lymph node metastasis of breast cancer using MRI and mammography.

OBJECTIVES: We aimed to develop a multi-modality model to predict axillary lymph node (ALN) metastasis by combining clinical predictors with radiomic features from magnetic resonance imaging (MRI) and mammography (MMG) in breast cancer. This model might potentially eliminate unnecessary axillary surgery in cases without ALN metastasis, thereby minimizing surgery-related complications. METHODS: We retrospectively enrolled 485 breast cancer patients from two hospitals and extracted radiomics features from tumor and lymph node regions on MRI and MMG images. After feature selection, three random forest models were built using the retained features, respectively. Significant clinical factors were integrated with these radiomics models to construct a multi-modality model. The multi-modality model was compared to radiologists' diagnoses on axillary ultrasound and MRI. It was also used to assist radiologists in making a secondary diagnosis on MRI. RESULTS: The multi-modality model showed superior performance with AUCs of 0.964 in the training cohort, 0.916 in the internal validation cohort, and 0.892 in the external validation cohort. It surpassed single-modality models and radiologists' ALN diagnosis on MRI and axillary ultrasound in all validation cohorts. Additionally, the multi-modality model improved radiologists' MRI-based ALN diagnostic ability, increasing the average accuracy from 70.70 to 78.16% for radiologist A and from 75.42 to 81.38% for radiologist B. CONCLUSION: The multi-modality model can predict ALN metastasis of breast cancer accurately. Moreover, the artificial intelligence (AI) model also assisted the radiologists to improve their diagnostic ability on MRI. CLINICAL RELEVANCE STATEMENT: The multi-modality model based on both MRI and mammography images allows preoperative prediction of axillary lymph node metastasis in breast cancer patients. With the assistance of the model, the diagnostic efficacy of radiologists can be further improved. KEY POINTS: • We developed a novel multi-modality model that combines MRI and mammography radiomics with clinical factors to accurately predict axillary lymph node (ALN) metastasis, which has not been previously reported. • Our multi-modality model outperformed both the radiologists' ALN diagnosis based on MRI and axillary ultrasound, as well as single-modality radiomics models based on MRI or mammography. • The multi-modality model can serve as a potential decision support tool to improve the radiologists' ALN diagnosis on MRI.

Consensus Clustering Analysis Based on Enhanced-CT Radiomic Features: Esophageal Squamous Cell Carcinoma patients' 3-Year Progression-Free Survival.

RATIONALE AND OBJECTIVES: To assess the efficacy of consensus cluster analysis based on CT radiomics in stratifying risk and predicting postoperative progression-free survival (PFS) in patients diagnosed with esophageal squamous cell carcinoma (ESC). MATERIALS AND METHODS: We conducted a retrospective study involving 546 patients diagnosed with ESC between January 2016 and March 2021. All patients underwent preoperative enhanced CT examinations. From the enhanced CT images, radiomics features were extracted, and a consensus clustering algorithm was applied to group the patients based on these features. Statistical analysis was performed to examine the relationship between the clustering results and gene protein expression, histopathological features, and patients' 3-year PFS. We applied the Kruskal-Wallis test for continuous data, chi-square or Fisher's exact tests for categorical data, and the log-rank test for PFS. RESULTS: This study identified four groups: Cluster 1 (n = 100, 18.3%), Cluster 2 (n = 197, 36.1%), Cluster 3 (n = 205, 37.5%), and Cluster 4 (n = 44, 8.1%). The cancer gene Breast Cancer Susceptibility Gene 1 (BRCA1) was most highly expressed in Cluster 4 (75%), showing significant differences between the four subtypes with a P-value of 0.035. The expression of programmed death-1 (PD-1) was highest in Cluster 1 (51%), with a P-value of 0.022. Vascular invasion occurred most frequently in Cluster 2 (28.9%), with a P-value of 0.022. The majority of patients with stage T3-4 were in Cluster 2 (67%), with a P-value of 0.003. Kaplan-Meier survival analysis revealed significant differences in PFS between the four groups (P = 0.013). Among them, patients in Cluster 1 had the best prognosis, while those in Cluster 2 had the worst. CONCLUSION: This study highlights the effectiveness of consensus clustering analysis based on enhanced CT radiomics features in identifying associations between radiomics features, histopathological characteristics, and prognosis in different clusters. These findings provide valuable insights for clinicians in accurately and effectively evaluating the prognosis of esophageal cancer.

Associating Peritoneal Metastasis With T2-Weighted MRI Images in Epithelial Ovarian Cancer Using Deep Learning and Radiomics: A Multicenter Study.

BACKGROUND: The preoperative diagnosis of peritoneal metastasis (PM) in epithelial ovarian cancer (EOC) is challenging and can impact clinical decision-making. PURPOSE: To investigate the performance of T2 -weighted (T2W) MRI-based deep learning (DL) and radiomics methods for PM evaluation in EOC patients. STUDY TYPE: Retrospective. POPULATION: Four hundred seventy-nine patients from five centers, including one training set (N = 297 [mean, 54.87 years]), one internal validation set (N = 75 [mean, 56.67 years]), and two external validation sets (N = 53 [mean, 55.58 years] and N = 54 [mean, 58.22 years]). FIELD STRENGTH/SEQUENCE: 1.5 or 3 T/fat-suppression T2W fast or turbo spin-echo sequence. ASSESSMENT: ResNet-50 was used as the architecture of DL. The largest orthogonal slices of the tumor area, radiomics features, and clinical characteristics were used to construct the DL, radiomics, and clinical models, respectively. The three models were combined using decision-level fusion to create an ensemble model. Diagnostic performances of radiologists and radiology residents with and without model assistance were evaluated. STATISTICAL TESTS: Receiver operating characteristic analysis was used to assess the performances of models. The McNemar test was used to compare sensitivity and specificity. A two-tailed P < 0.05 was considered significant. RESULTS: The ensemble model had the best AUCs, outperforming the DL model (0.844 vs. 0.743, internal validation set; 0.859 vs. 0.737, external validation set I) and clinical model (0.872 vs. 0.730, external validation set II). After model assistance, all readers had significantly improved sensitivity, especially for those with less experience (junior radiologist1, from 0.639 to 0.820; junior radiologist2, from 0.689 to 0.803; resident1, from 0.623 to 0.803; resident2, from 0.541 to 0.738). One resident also had significantly improved specificity (from 0.633 to 0.789). DATA CONCLUSIONS: T2W MRI-based DL and radiomics approaches have the potential to preoperatively predict PM in EOC patients and assist in clinical decision-making. EVIDENCE LEVEL: 4 TECHNICAL EFFICACY: Stage 2.

Deep Learning Radiomics Nomogram Based on Magnetic Resonance Imaging for Differentiating Type I/II Epithelial Ovarian Cancer.

RATIONALE AND OBJECTIVES: To develop and validate a T2-weighted magnetic resonance imaging (MRI)-based deep learning radiomics nomogram (DLRN) to differentiate between type I and type II epithelial ovarian cancer (EOC). MATERIALS AND METHODS: This multicenter study incorporated 437 patients from five centers, divided into training (n = 271), internal validation (n = 68), and external validation (n = 98) sets. The deep learning (DL) model was constructed using the largest orthogonal slices of the tumor area. The extracted radiomics features were employed in building the radiomics model. The clinical model was developed based on clinical characteristics. A DLRN was built by integrating the DL signature, radiomics signature, and independent clinical predictors. Model performances were evaluated through receiver operating characteristic (ROC) analysis, Brier score, calibration curve, and decision curve analysis (DCA). The areas under the ROC curve (AUCs) were compared using the DeLong test. A two-tailed P < 0.05 was considered significantly different. RESULTS: The DLRN exhibited satisfactory discrimination between type I and type II EOC with the AUC of 0.888 (95% confidence interval [CI] 0.810, 0.966) and 0.866 (95% CI 0.786, 0.946) in the internal and external validation sets, respectively. These AUCs significantly exceeded those of the clinical model (P = 0.013 and 0.043, in the internal and external validation sets, respectively). The DLRN demonstrated optimal classification accuracy and clinical application value, according to Brier scores, calibration curves, and DCA. CONCLUSION: A T2-weighted MRI-based DLRN showed promising potential in differentiating between type I and type II EOC, which could offer assistance in clinical decision-making.

Developing and validating a nomogram based on skeletal muscle index and clinical scoring system for prediction of liver failure after hepatectomy.

Background and objectives: Hepatectomy is the preferred treatment for patients with liver tumors. Post-hepatectomy liver failure (PHLF) remains one of the most fatal postoperative complications. We aim to explore the risk factors of PHLF and create a nomogram for early prediction of PHLF. Methods: We retrospectively analyzed patients undergoing hepatectomy at the Affiliated Huaian No. 1 People's Hospital of Nanjing Medical University between 2015 and 2022, and the patients were divided into training and internal validation cohorts at an 8:2 ratio randomly. The patients undergoing liver resection from the Affiliated Huaian Hospital of Xuzhou Medical University worked as external validation. Then, a nomogram was developed which was based on multivariate analyses to calculate the risk of PHLF. The area under the ROC curve (AUROC) and Hosmer -Lemeshow test was used to evaluate the prediction effect of the model. Results: A total of 421 eligible patients were included in our study. Four preoperative variables were identified after multivariate analysis as follows, ASA (American Society of Anesthesiologists) score, Child-Pugh score, SMI (Skeletal muscle index), and MELD (Model for end-stage liver disease) score as independent predictors of PHLF. The area under the ROC curve of the predictive model in the training, internal, and external validation cohorts were 0.89, 0.82, and 0.89. Hosmer -Lemeshow P values in the training, internal, and external validation cohorts were 0.91, 0.22, and 0.15. The Calibration curve confirmed that our nomogram prediction results were in accurate agreement with the actual occurrence of PHLF. Conclusion: We construct a nomogram to predict the grade B/C PHLF of ISGLS (International Study Group of Liver Surgery) in patients who underwent hepatic resection based on risk factors. This tool can provide a visual and accurate preoperative prediction of the grade B/C PHLF and guide the next step of clinical decision-making.

Predictive value of Gd-EOB-DTPA -enhanced magnetic resonance imaging for post-hepatectomy liver failure: a systematic review and meta-analysis.

BACKGROUND: Accurate preoperative diagnosis of post-hepatectomy liver failure (PHLF) is particularly important to improve the prognosis of patients. PURPOSE: To evaluate the predictive value of Gd-EOB-DTPA-enhanced magnetic resonance imaging (MRI) for post-hepatectomy liver failure. MATERIAL AND METHODS: A systematic search was performed in the PubMed, Embase, the Cochrane Library, and Web of Science databases to find relevant original articles published up to December 2021. The included studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. The bivariate random-effects model was used to assess the diagnostic authenticity. Meta-regression analyses were performed to analyze the potential heterogeneity. RESULTS: In total, 13 articles were included. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the area under the summary receiver operating characteristic curves were 88% (95% confidence interval [CI] = 0.80-0.94), 80% (95% CI = 0.73-0.86), 4.4 (95% CI = 3.3-5.9), 0.14 (95% CI = 0.08-0.25), 31 (95% CI = 17-57), and 0.91 (95% CI = 0.89-0.94), respectively. There was no publication bias and threshold effect in our study. CONCLUSION: Gd-EOB-DTPA-enhanced MRI is a potentially useful for the prediction of PHLF after major hepatectomy.

Correlative investigation between routine clinical parameters of dual-energy computed tomography and the outcomes of extracorporeal shock wave lithotripsy in children with urolithiasis: a retrospective study.

PURPOSE: To evaluate the associations of DECT parameters with extracorporeal shock wave lithotripsy (ESWL) outcomes in pediatric patients. METHODS: A retrospective study of consecutive patients with calculi who underwent ESWL and DECT in our hospital was performed in 2011-2019. The primary outcome was DECT imaging's correlation with ESWL outcomes. The secondary outcome was to determine DECT parameters independently predicting ESWL outcomes, including stone-free (SF) and residual stone (RS) statuses. RESULTS: The study included 207 patients. The mean CT attenuations at 140 kVp, 80 kVp, and 120 kVp and effective atomic number (Zeff) were significantly correlated with stone free (SF) and residual stone (RS) (P < 0.05). Areas under the curves (AUCs) of CT attenuations at 120 kVp, 80 kVp, 140 kVp, and dual-energy index (DEI) were 0.784 (95% CI 0.672-0.897), 0.780 (95% CI 0.677-0.884), 0.766 (95% CI 0.658-0.885), and 0.709 (95% CI 0.578-0.840) (all P < 0.05). With cutoffs of 882.5, 1330.5, 1042.5, and 0.103 for CT attenuations at 140 kVp, 80 kVp, 120 kVp, and DEI, respectively, sensitivities and specificities were 75.0% and 31.1%, 83.3% and 31.8%, 80.3% and 31.1%, and 58.3% and 44.7%, respectively. CONCLUSION: Our results demonstrated that the parameters of DECT could be used to predict ESWL outcomes (especially the SF status) in children with urolithiasis. ESWL success can be accurately predicted by DECT, and it is hard to predict ESWL failure.

The status and characteristics of urinary stone composition in China.

OBJECTIVES: To explore characteristics of urinary stone composition in China, and determine the effects of gender, age, body mass index (BMI), stone location, and geographical region on stone composition. PATIENTS AND METHODS: We prospectively used Fourier-transform infrared spectroscopy to analyse stones from consecutive patients presenting with new-onset urolithiasis at 46 hospitals in seven geographical areas of China, between 1 June 2010 and 31 May 2015. Chi-squared tests and logistic regression analyses were used to determine associations between stone composition and gender, age, BMI, stone location, and geographical region. RESULTS: The most common stone constituents were: calcium oxalate (CaOx; 65.9%), carbapatite (15.6%), urate (12.4%), struvite (2.7%), and brushite (1.7%). CaOx and urate stones occurred more frequently in males, whereas carbapatite and struvite were more common in females (P < 0.01). CaOx and carbapatite were more common in those aged 30-50 and 20-40 years than in other groups. Brushite and struvite were most common amongst those aged <20 and >70 years. The detection rate of urate increased with age, whilst cystine decreased with age. Obese patients were more likely to have urate stones than carbapatite or brushite stones (P < 0.01). CaOx, carbapatite, brushite, and cystine stones were more frequently found in the kidney than other types, whereas urate and struvite were more frequent in the bladder (P < 0.01). Stone composition varied by geographical region. CONCLUSIONS: The most common stone composition was CaOx, followed by carbapatite, urate, struvite, and brushite. Stone composition differed significantly in patients grouped by gender, age, BMI, stone location, and geographical region.

Acute kidney injury caused by ceftriaxone-induced urolithiasis in children: a single-institutional experience in diagnosis, treatment and follow-up.

PURPOSE: To evaluate our clinical outcomes in managing acute kidney injury (AKI) resulting from ceftriaxone-induced urolithiasis with emergency treatment. METHODS: From July 2008 to July 2013, a series of 15 patients including 12 males and 3 females were admitted to our center. The mean age of them was (4.76 ± 3.74) years. A chief complaint of anuria was presented in 12 (80.0 %) patients for 13 h-4 days and that of oliguria in three (20.0 %) patients for 20 h-10 days. All of them were diagnosed of postrenal AKI resulting from ceftriaxone-induced urolithiasis and underwent emergency hospitalization. RESULTS: Double-J stenting with cystoscopy was successfully performed in nine patients (60.0 %), and ureteroscopy was applied in four patients (26.7 %). One patient (6.7 %) underwent unilateral double-J insertion combined with contralateral percutaneous nephrostomy, and one (6.7 %) underwent open surgery. Loose texture and sandlike stones, the main characteristics of these stones, made them excreted spontaneously after the initial treatment, whereas only one patient (6.7 %) underwent additional ureterolithotomy due to many residual calculi. Serum creatinine and blood urea nitrogen recovered to normal levels within 3 days. All specimens were collected and analyzed by infrared spectrum, with results demonstrating that the main composition was ceftriaxone calcique. All patients were followed up for 11 months-5 years (mean 33.80 ± 22.56 months). No one turned to irreversible renal failure. CONCLUSIONS: Ceftriaxone could result in urolithiasis in children, which could also cause AKI. Appropriate and timely surgical management by conventional treatments will mostly lead to full recovery of renal functions.

Extracorporeal shock wave lithotripsy is effective in treating single melamine induced urolithiasis in infants and young children.

PURPOSE: We evaluated the safety and efficacy of extracorporeal shock wave lithotripsy in the treatment of single melamine induced urolithiasis in infants and young children. MATERIALS AND METHODS: A total of 189 infants and young children with single melamine induced urolithiasis were referred to our center for treatment with extracorporeal shock wave lithotripsy between March 2009 and July 2010. Location of the calculus was proximal ureteral in 17 patients, mid ureteral in 5, distal ureteral in 26 and kidney in 141. Stone size ranged from 3.8 to 25 mm (mean ± SD 9.79 ± 3.83). RESULTS: All patients underwent extracorporeal shock wave lithotripsy using the same device with an energy ranging from 8 to 12 kV. Stone-free rate was 97.88%, clinically insignificant residual fragment rate was 1.59% and repeat treatment rate was 2.65%. A total of 180 patients (95.24%) required only 1 lithotripsy session and 5 (2.65%) required 2 sessions. Mean ± SD number of shock waves delivered per session was 580.36 ± 190.69 (range 65 to 950). Extracorporeal shock wave lithotripsy failed to fragment stones in only 1 infant, who had a proximal ureteral stone. A total of 181 specimens were collected and analyzed by infrared spectrum, with results demonstrating that the main composition was uric acid and melamine. All patients were followed for a mean of 28 months (range 20 to 36). No severe complication, such as renal subcapsular hemorrhage, hypertension, kidney rupture or lung injury, was observed. CONCLUSIONS: Extracorporeal shock wave lithotripsy with low energy can effectively disintegrate melamine induced calculi. This approach has become our preferred method for treating single melamine induced urolithiasis in infants and young children.

Treatment of renal stones in infants: comparing extracorporeal shock wave lithotripsy and mini-percutaneous nephrolithotomy.

The objective of the study is to compare the efficacy and safety of extracorporeal shock wave lithotripsy (ESWL) and mini-percutaneous nephrolithotomy (MPCNL) in treating renal stones sizing 15-25 mm in infants <3 years. Forty-six infants with renal stones sizing 15-30 mm were treated by either ESWL (22 renal units in 22 infants) using Dornier compact delta lithotripter or MPCNL (25 renal units in 24 infants) using 14F-18F renal access under general anesthesia. The operation time, stone-free rate, re-treatment rate, and complications between the two groups were compared with the χ(2), Mann-Whitney U, and Student's t tests. No significant differences in mean age and stone size were observed between the two groups. The 1- and 3-month postoperative stone-free rates were 84 and 96% in MPCNL group and were 31.8 and 86.4% in ESWL group. The re-treatment and complication rates were significantly higher in ESWL group than in MPCNL group (50 vs. 12%, P = 0.004; 16.0 vs. 45.5%, P = 0.028). The stone recurrence rate was similar between the two groups. No significant changes of serum creatinine (Cr) level and glomerular filtration rate were observed in both groups. In conclusion, MPCNL is an effective and feasible alternative monotherapy for large renal stones (15-25 mm) in infants, with a higher stone-free rate and a lower complication rate when compared with ESWL.

Occurrence and ordination of dichlorodiphenyltrichloroethane and hexachlorocyclohexane in agricultural soils from Guangzhou, China.

On a global scale tropical regions in developing countries are thought to be significant source areas of organochlorine pesticides (OCPs), owing to a long history of widespread use and only a recent production ban or restriction on the application of these pesticides. In the present study, 32 soil samples were collected in 2004 from agriculture lands around the urban area of Guangzhou, in southern China, and analyzed for residues of OCPs including p,p'-DDT, p,p'-DDE, p,p'-DDD, and alpha-, beta-, gamma-, and delta-HCH. The dry weight concentrations of SigmaHCH (SigmaHCH = alpha-HCH + beta-HCH + gamma-HCH + delta-HCH) ranged from 0.2 to 103.9 ng/g, with a median of 4.4 ng/g. Residues of SigmaDDT (SigmaDDT = p,p'-DDT + p,p'-DDE + p,p'-DDD) ranged from 7.6 to 662.9 ng/g, with a median of 67.3 ng/g. The predominance of beta-HCH among HCHs in most soil samples suggested that they were from historical contamination rather than recent input. The mean HCH alpha/gamma-ratio of 2.72 was lower than that of technical HCHs, possibly due to more loss of alpha-HCH via evaporation from soil with time, conversion of gamma-HCH to alpha-HCH or recent application of lindane in the region. The mean ratio of (DDE + DDD)/SigmaDDT was 0.54, indicating that quite a portion of DDT in soils was degraded since its official ban in 1983. Higher DDT concentrations with lower (DDE + DDD)/SigmaDDT ratios at a few sites suggested possible local DDT sources via the application of Dicofol. A positive but weak correlation (r = 0.449, p < 0.01) between DDT residues and TOC contents implied that soil organic matter might enhance adsorption of DDT in soils in the tropical regions. Hierarchical cluster analysis and principal component analysis were also performed to study the distribution and compositional patterns of OCPs as well as their sources and environmental fates within the study area.