The role of 18F-FDG PET/CT in primary cutaneous lymphoma: an educational review

Computed tomography-based delta-radiomics analysis for preoperative prediction of ISUP pathological nuclear grading in clear cell renal cell carcinoma

BACKGROUND

Nuclear grading of clear cell renal cell carcinoma (ccRCC) plays a crucial role in diagnosing and managing the disease.

OBJECTIVE

To develop and validate a CT-based Delta-Radiomics model for preoperative assessment of nuclear grading in renal clear cell carcinoma.

MATERIALS AND METHODS

This retrospective analysis included surgical cases of 146 ccRCC patients from two medical centers from December 2018 to December 2023, with 117 patients from Hospital and 29 from the *Hospital Affiliated to University of **. Radiomic features were extracted from whole-abdomen CT images, and the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm was used for feature selection. The Multi-Layer Perceptron (MLP) approach was employed to construct five predictive models (RAD_NE, RAD_AP, RAD_VP, RAD_Delta1, RAD_Delta2). The models were evaluated using area under the curve (AUC), accuracy, sensitivity, and specificity, while clinical utility was assessed through Decision Curve Analysis (DCA).

RESULTS

A total of 1,834 radiomic features were extracted from the three phases of the CT images for each model. The models demonstrated strong classification performance, with AUC values ranging from 0.837 to 0.911 in the training set and 0.608 to 0.869 in the test set. The Rad_Delta1 and Rad_Delta2 models demonstrated advantages in predicting ccRCC pathological grading.The AUC value of the Rad_Delta1 is 0.911in the training set and 0.771 in the external verifcation set.The AUC value of the Rad_Delta2 is 0.881 in the training set and0.608 in the external verifcation set. DCA curves confirmed the clinical applicability of these models.

CONCLUSION

CT-based delta-radiomics shows potential in predicting the pathological grading of clear cell renal cell carcinoma (ccRCC).

Identification of Skin Lesions by Snapshot Hyperspectral Imaging

This study pioneers the application of artificial intelligence (AI) and hyperspectral imaging (HSI) in the diagnosis of skin cancer lesions, particularly focusing on Mycosis fungoides (MF) and its differentiation from psoriasis (PsO) and atopic dermatitis (AD). By utilizing a comprehensive dataset of 1659 skin images, including cases of MF, PsO, AD, and normal skin, a novel multi-frame AI algorithm was used for computer-aided diagnosis. The automatic segmentation and classification of skin lesions were further explored using advanced techniques, such as U-Net Attention models and XGBoost algorithms, transforming images from the color space to the spectral domain. The potential of AI and HSI in dermatological diagnostics was underscored, offering a noninvasive, efficient, and accurate alternative to traditional methods. The findings are particularly crucial for early-stage invasive lesion detection in MF, showcasing the model's robust performance in segmenting and classifying lesions and its superior predictive accuracy validated through k-fold cross-validation. The model attained its optimal performance with a k-fold cross-validation value of 7, achieving a sensitivity of 90.72%, a specificity of 96.76%, an F1-score of 90.08%, and an ROC-AUC of 0.9351. This study marks a substantial advancement in dermatological diagnostics, thereby contributing significantly to the early and precise identification of skin malignancies and inflammatory conditions.