Objective:To explore the value of the Nomogram model established by CT radiomics combined with clinical indicators for prediction of the severity of early acute pancreatitis (AP).Methods:From January 2016 to March 2023, the AP patients in the Second Affiliated Hospital of Soochow University were retrospectively collected. According to the revised Atlanta classification and definition of acute pancreatitis in 2012, all patients were divided into the severe group and the non-severe group. All patients were first diagnosed, and abdominal CT plain scan and enhanced scan were completed within 1 week. Patients were randomly (random number) divided into training and validation groups at a ratio of 7:3. The pancreatic parenchyma was delineated as the region of interest on each phase CT images, and the radiomics features were extracted by python software. LASSO regression and 10-fold cross-validation were used to reduce the dimension and select the optimal features to establish the radiomics signature. Multivariate Logistic regression was used to select the independent predictors of severe acute pancreatitis (SAP), and a clinical model was established. A Nomogram model was established by combining CT radiomics signature and clinical independent predictors. Receiver operating characteristic (ROC) curve and decision curve analysis (DCA) were used to evaluate the predictive efficacy of each model.Results:Total of 205 AP patients were included (59 cases in severe group, 146 cases in non-severe group). 3, 5, 5 and 5 optimal radiomics features were selected from the plain CT scan, arterial phase, venous phase and delayed phase images of all patients, and the radiomics models were established. Among them, the arterial phase radiomics model had relatively better performance in predicting SAP, with an area under curve (AUC) of 0.937 in the training group and 0.913 in the validation group. Multivariate Logistic regression showed that C-reactive protein (CRP) and lactate dehydrogenase (LDH) were independent predictors of SAP, and they were used to establish a clinical model. The AUC in the training and validation groups were 0.879 and 0.889, respectively. The Nomogram model based on arterial phase CT radiomics signature, CRP and LDH was established, and the AUC was 0.956 and 0.947 in the training group and validation group, respectively. DCA showed that the net benefit of Nomogram model was higher than that of clinical model or radiomics model alone.Conclusions:The Nomogram model established by CT radiomics combined with clinical indicators has high application value for early prediction of the severity of AP, which is conducive to the formulation of clinical treatment plans and prognosis evaluation.

Objective:To explore the value of machine learning models based on multiple structural MRI features for diagnosis of Parkinson disease (PD).Methods:The clinical and imaging data of 60 PD patients (PD group) diagnosed in the Neurology Department of the Second Affiliated Hospital of Soochow University from November 2017 to August 2019 and 56 normal elderly people (NC group) recruited from the community were retrospectively analyzed. All subjects underwent brain MR imaging. Multiple structural MRI features were extracted from cerebellum, deep nuclei and of brain cortex based on different partition templates. The Mann-Whitney U test, as well as least absolute shrinkage and selection operator regression were used to select the most discriminating features. Finally, logistic regression (LR) and linear discriminant analysis (LDA) classifier combined with the 5-fold cross-validation scheme were used to construct the models based on structural features of cerebellum, deep nuclei and cortex, and a combined model based on all features. The receiver operating characteristic curves were drawn, and the diagnostic performance and clinical net benefit of each model were evaluated by the area under curve (AUC) and the decision curve analysis (DCA). Results:In total, four cerebellum (asymmetry index of Lobule Ⅵ volume, asymmetry index of Lobule ⅦB cortical thickness, asymmetry index of total gray matter volume and absolute value of right Lobule Ⅵ gray matter volume), 3 deep nuclei (absolute value of right nucleus accumbens volume, absolute and relative value of total nucleus accumbens volume) and 3 cortex features (local gyration index of left PFm, local fractal dimension of right superior frontal gyrus and sulcal depth of left superior occipital gyrus) were selected as the most discriminating features, and the related models were constructed. In validation set, the AUC of cerebellum, deep nuclei, cortex and combined models for diagnosis of PD based on LR classifier were 0.692, 0.641, 0.747 and 0.816; the AUC of cerebellum, deep nuclei, cortex and combined models for diagnosis of PD based on LDA classifier were 0.726, 0.610, 0.752 and 0.818. The diagnostic efficiency of the combined models based on LR and LDA classifiers were significantly better than those of other models ( P<0.05). The DCA curve demonstrated that the combined models based on LR and LDA classifiers showed the highest clinical net benefit. Conclusion:The combined models with all structural features of cerebellum, deep nuclei and cortex included based on LR and LDA classifiers showed favorable performance and clinical net benefit for diagnosis of PD, which have the potential application value in clinical diagnosis.

Objective To construct the machine learning models based on the radiomic features of non-contrast and enhanced CT and to evaluate the predictive value in the risk stratification of gastrointestinal stromal tumor(GIST).Methods A total of 182 patients with pathologically confirmed GIST were randomly divided into a training set and a validation set at a ratio of 7∶3.The volume of interest(VOI)was outlined in the non-contrast phase,arterial phase and venous phase,and its radiomic features were extracted.The most valuable radiomic features were selected using the least absolute shrinkage and selection operator(LASSO)algorithm.The logistic regression(LR)classifier was used to construct the prediction models based on single-phase or multi-phase images.The predictive efficacy of the different models was compared by using receiver operating characteristic(ROC)curves.Results Four,three,and four radiomic features were selected in the non-contrast phase,arterial phase and venous phase,and 4 models were constructed in total.Among the single-phase models,the venous phase had better predictive efficacy,with the area under the curve(AUC)of 0.932[95％confidence interval(CI)0.873-0.969]and 0.924(95％CI 0.819-0.979)in the training and validation sets.The predictive efficacy of the combined model was improved,with the AUC of 0.946(95％CI 0.891-0.978)and 0.938(95％CI 0.838-0.986).Conclusion The venous phase model can predict the risk stratification of GIST accurately,and the prediction efficacy can be improved by combining the non-contrast and arterial phases.

Objective:To investigate the predictive value of clinical factor model, deep learning model based on baseline plain CT images, and combination of both for predicting hematoma expansion in cerebral hemorrhage.Methods:The study was cross-sectional. Totally 471 cerebral hemorrhage patients who were firstly diagnosed in the Second Affiliated Hospital of Soochow University from January 2017 to December 2021 were collected retrospectively. These patients were randomly divided into a training dataset ( n=330) and a validation dataset ( n=141) at a ratio of 7∶3 by using the random function. All patients underwent two noncontrast CT examinations within 24 h and an increase in hematoma volume of >33% or an absolute increase in hematoma volume of >6 ml was considered hematoma enlargement. According to the presence or absence of hematoma enlargement, all patients were divided into hematoma enlargement group and hematoma non-enlargement group.Two-sample t test, Mann-Whitney U test or χ2 test were used for univariate analysis. The factors with statistically significant differences were included in multivariate logistic regression analysis, and independent influences related to hematoma enlargement were screened out to establish a clinical factor model. ITK-SNAP software was applied to manually label and segment the cerebral hemorrhage lesions on plain CT images to train and build a deep learning model based on ResNet50 architecture. A combination model for predicting hematoma expansion in cerebral hemorrhage was established by combining independent clinical influences with deep learning scores. The value of the clinical factor model, the deep learning model, and the combination model for predicting hematoma expansion in cerebral hemorrhage was evaluated using receiver operating characteristic (ROC) curves and decision curves in the training and validation datasets. Results:Among 471 cerebral hemorrhage patients, 136 cases were in the hematoma enlargement group and 335 cases were in the hematoma non-enlargement group. Regression analyses showed that male ( OR=1.790, 95% CI 1.136-2.819, P=0.012), time of occurrence ( OR=0.812, 95% CI 0.702-0.939, P=0.005), history of oral anticoagulants ( OR=2.157, 95% CI 1.100-4.229, P=0.025), admission Glasgow Coma Scale score ( OR=0.866, 95% CI 0.807-0.929, P<0.001) and red blood cell distribution width ( OR=1.045, 95% CI 1.010-1.081, P=0.011) were the independent factors for predicting hematoma expansion in cerebral hemorrhage. ROC curve analysis showed that in the training dataset, the area under the curve (AUC) of clinical factor model, deep learning model and combination model were 0.688 (95% CI 0.635-0.738), 0.695 (95% CI 0.642-0.744) and 0.747 (95% CI 0.697-0.793) respectively. The AUC of the combination model was better than that of the clinical model ( Z=0.54, P=0.011) and the deep learning model ( Z=2.44, P=0.015). In the validation dataset, the AUC of clinical factor model, deep learning model and combination model were 0.687 (95% CI 0.604-0.763), 0.683 (95% CI 0.599-0.759) and 0.736 (95% CI 0.655-0.806) respectively, with no statistical significance. Decision curves showed that the combination model had the highest net benefit rate and strong clinical practicability. Conclusions:Both the deep learning model and the clinical factor model established in this study have some predictive value for hematoma expansion in cerebral hemorrhage; the combination model established by the two together has the highest predictive value and can be applied to predict hematoma expansion.

A 3-year-old male patient was diagnosed with neurofibromatosis type 1(NF1) for two years. The patient has multiple neurofibromas in retroperitoneum, lumbococcygeal paravertebral, lumbosacral spinal canal, and foramina. Due to retroperitoneal mass compression, the child suffered from urological complications such as hydronephrosis, ureterdilation, neurogenic bladder, etc., which seriously affected the urination function and resulted in multiple surgical treatments. Currently, the patient has been treated with mitogen activates extracelluar signal-regulated kinases(MEK) inhibitor selumetinib targeted therapy, and has voluntarily urinated, and his general state is better than before medication. The diagnosis and treatment of this case reflects the importance of multidisciplinary collaboration in the diagnosis and treatment of rare diseases.