Machine Learning-Based Radiological Features and Diagnostic Predictive Model of Xanthogranulomatous Cholecystitis

Deep-learning models for differentiation of xanthogranulomatous cholecystitis and gallbladder cancer on ultrasound

BACKGROUND

The radiological differentiation of xanthogranulomatous cholecystitis (XGC) and gallbladder cancer (GBC) is challenging yet critical. We aimed at utilizing the deep learning (DL)-based approach for differentiating XGC and GBC on ultrasound (US).

METHODS

This single-center study comprised consecutive patients with XGC and GBC from a prospectively acquired database who underwent pre-operative US evaluation of the gallbladder lesions. The performance of state-of-the-art (SOTA) DL models (GBCNet-convolutional neural network [CNN] and RadFormer, transformer) for XGC vs. GBC classification in US images was tested and compared with popular DL models and a radiologist.

RESULTS

Twenty-five patients with XGC (mean age, 57 ± 12.3, 17 females) and 55 patients with GBC (mean age, 54.6 ± 11.9, 38 females) were included. The performance of GBCNet and RadFormer was comparable (sensitivity 89.1% vs. 87.3%, p = 0.738; specificity 72% vs. 84%, p = 0.563; and AUC 0.744 vs. 0.751, p = 0.514). The AUCs of DenseNet-121, vision transformer (ViT) and data-efficient image transformer (DeiT) were significantly smaller than of GBCNet (p = 0.015, 0.046, 0.013, respectively) and RadFormer (p = 0.012, 0.027, 0.007, respectively). The radiologist labeled US images of 24 (30%) patients non-diagnostic. In the remaining patients, the sensitivity, specificity and AUC for GBC detection were 92.7%, 35.7% and 0.642, respectively. The specificity of the radiologist was significantly lower than of GBCNet and RadFormer (p = 0.001).

CONCLUSION

SOTA DL models have a better performance than radiologists in differentiating XGC and GBC on the US.

Gallbladder cancer masquerading as xanthogranulomatous cholecystitis: a case report and literature review

Xanthogranulomatous cholecystitis (XGC) is a rare type of cholecystitis that, despite being benign poses diagnostic challenges due to its low prevalence and need for consensus on diagnostic criteria. Consequently, distinguishing XGC from gallbladder cancer (GBC) is challenging, leading to clinical misdiagnoses. This article presents a case where a patient initially diagnosed with GBC was later found to have XGC.

Machine learning-based diagnostic model for preoperative differentiation between xanthogranulomatous cholecystitis and gallbladder carcinoma: a multicenter retrospective cohort study

Background

Xanthogranulomatous cholecystitis (XGC) and gallbladder carcinoma (GBC) share similar imaging and serological profiles, posing significant challenges in accurate preoperative diagnosis. This study aimed to identify reliable indicators and develop a predictive model to differentiate between XGC and GBC.

Methods

This retrospective study involved 436 patients from Zhejiang Provincial People's Hospital and The Affiliated Lihuili Hospital of Ningbo University. Comprehensive preoperative imaging, including ultrasound, Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and blood tests, were analyzed. Machine learning (Random Forest method) was employed for variable selection, and a multivariate logistic regression analysis was used to construct a nomogram for predicting GBC. Statistical analyses were performed using SPSS and RStudio software.

Results

The study identified gender, Murphy's sign, absolute neutrophil count, glutamyl transpeptidase level, carcinoembryonic antigen level, and comprehensive imaging diagnosis as potential risk factors for GBC. A nomogram incorporating these factors demonstrated high predictive accuracy for GBC, outperforming individual or combined traditional diagnostic methods. External validation of the nomogram showed consistent results.

Conclusion

The study successfully developed a predictive nomogram for distinguishing GBC from XGC with high accuracy. This model, integrating multiple clinical and imaging indicators, offers a valuable tool for clinicians in making informed diagnostic decisions. The findings advocate for the use of comprehensive preoperative evaluations combined with advanced analytical tools to improve diagnostic accuracy in complex medical conditions.

Preoperative differential diagnosis of xanthogranulomatous cholecystitis and gallbladder carcinoma

BACKGROUND

Because the clinical and imaging features of xanthogranu-lomatous cholecystitis (XGC) and gallbladder carcinoma (GBC) are very similar, it is often difficult to distinguish them clinically. Based on abdominal ultrasound and contrast-enhanced CT, this study analyzed the differences in demographic characteristics, laboratory indicators, tumor markers, and imaging features between patients with XGC and those with GBC. Then a diagnostic model was constructed to assist clinicians in the diagnosis and treatment of the two conditions.

AIM

To analyze the differences in clinical and imaging features between XGC and GBC patients, and to establish a diagnostic model for the two conditions.

METHODS

From January 2011 to September 2022, 67 patients with XGC and 139 patients with GBC who underwent abdominal ultrasonography and abdominal contrast-enhanced CT and had definite postoperative pathological diagnosis at Zhejiang Provincial People's Hospital were retrospectively analyzed. The differences in clinical manifestations and laboratory and imaging findings between the two groups were analyzed.

RESULTS

Gender, γ-glutamyl transpeptadase (GGT), carcinoembryonic antigen (CEA), mean gallbladder wall thickness, gallbladder wall thickening pattern (gallbladder wall involvement < 50%), gallstones, and retroperitoneal lymphadenopathy were independent risk factors for differentiating XGC from GBC. The cut-off values for GGT and CEA were 28 U/L and 3.2 ug/L, respectively.

CONCLUSION

There are significant differences in some clinical and imaging features between XGC and GBC, which can provide reference value for their preoperative differential diagnosis.

Development and validation of a prediction model based on machine learning algorithms for predicting the risk of heart failure in middle-aged and older US people with prediabetes or diabetes

BACKGROUND

The purpose of this study was to develop and validate a machine learning (ML) based prediction model for the risk of heart failure (HF) in patients with prediabetes or diabetes.

METHODS

We used 3527 subjects aged 40 years and older with a prior diagnosis of prediabetes or diabetes from the National Health and Nutrition Examination Survey (NHANES) from 2007 to 2018. The search for independent risk variables linked to HF was conducted using univariate and multivariable logistic regression analysis. The 3527 subjects were randomly divided into training set and validation set in a 7:3 ratio. Five ML models were built on the training set using five ML algorithms, including random forest (RF), and then validated on the validation set. Receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis and Bootstrap resampling method were used to measure the predictive performance of the five ML models.

RESULTS

Multivariate logistic regression analysis showed that age, poverty-to-income ratio, myocardial infarction condition, coronary heart disease condition, chest pain condition, and glucose-lowering medication use were independent predictors of HF. By comparing the performance of the five ML models, the RF model (AUC = 0.978) was the best prediction model.

CONCLUSIONS

The risk of HF in middle-aged and elderly patients with prediabetes or diabetes can be accurately predicted using ML models. The best prediction performance is presented by RF model, which can assist doctors in making clinical decisions.

Differential diagnoses of gallbladder tumors using CT-based deep learning

Background

The differential diagnosis between gallbladder cancer (GBC) and xanthogranulomatous cholecystitis (XGC) remains quite challenging, and can possibly lead to improper surgery. This study aimed to distinguish between XGC and GBC by combining computed tomography (CT) images and deep learning (DL) to maximize the therapeutic success of surgery.

Methods

We collected a dataset, including preoperative CT images, from 28 cases of GBC and 21 XGC patients undergoing surgery at our facility. It was subdivided into training and validation (n = 40), and test (n = 9) datasets. We built a CT patch-based discriminating model using a residual convolutional neural network and employed 5-fold cross-validation. The discriminating performance of the model was analyzed in the test dataset.

Results

Of the 40 patients in the training dataset, GBC and XGC were observed in 21 (52.5%), and 19 (47.5%) patients, respectively. A total of 61 126 patches were extracted from the 40 patients. In the validation dataset, the average sensitivity, specificity, and accuracy were 98.8%, 98.0%, and 98.5%, respectively. Furthermore, the area under the receiver operating characteristic curve (AUC) was 0.9985. In the test dataset, which included 11 738 patches, the discriminating accuracy for GBC patients after neoadjuvant chemotherapy (NAC) (n = 3) was insufficient (61.8%). However, the discriminating model demonstrated high accuracy (98.2%) and AUC (0.9893) for cases other than those receiving NAC.

Conclusion

Our CT-based DL model exhibited high discriminating performance in patients with GBC and XGC. Our study proposes a novel concept for selecting the appropriate procedure and avoiding unnecessary invasive measures.