Vision transformer and explainable transfer learning models for auto detection of kidney cyst, stone and tumor from CT-radiography

Abstract

Renal failure, a public health concern, and the scarcity of nephrologists around the globe have necessitated the development of an AI-based system to auto-diagnose kidney diseases. This research deals with the three major renal diseases categories: kidney stones, cysts, and tumors, and gathered and annotated a total of 12,446 CT whole abdomen and urogram images in order to construct an AI-based kidney diseases diagnostic system and contribute to the AI community’s research scope e.g., modeling digital-twin of renal functions. The collected images were exposed to exploratory data analysis, which revealed that the images from all of the classes had the same type of mean color distribution. Furthermore, six machine learning models were built, three of which are based on the state-of-the-art variants of the Vision transformers EANet, CCT, and Swin transformers, while the other three are based on well-known deep learning models Resnet, VGG16, and Inception v3, which were adjusted in the last layers. While the VGG16 and CCT models performed admirably, the swin transformer outperformed all of them in terms of accuracy, with an accuracy of 99.30 percent. The F1 score and precision and recall comparison reveal that the Swin transformer outperforms all other models and that it is the quickest to train. The study also revealed the blackbox of the VGG16, Resnet50, and Inception models, demonstrating that VGG16 is superior than Resnet50 and Inceptionv3 in terms of monitoring the necessary anatomy abnormalities. We believe that the superior accuracy of our Swin transformer-based model and the VGG16-based model can both be useful in diagnosing kidney tumors, cysts, and stones.

Theranostic roles of machine learning in clinical management of kidney stone disease

“…Using this approach, the urinary tract is detected by the first CNN model, whereas the stones are detected by the second CNN model [20] . Additionally, a total of six models have been designed and deployed using CT image datasets of kidney stones, cysts and tumors [50] . Both deep learning techniques (VGG16, Inceptionv3 and Resnet50) and Visual Transformer variants (EANet, CCT and Swin transformer algorithms) can be applied to differentiate KSD from renal cysts and tumors with 99.30 % accuracy achieved by Swin transformer-based model [50] .…”

“…Additionally, a total of six models have been designed and deployed using CT image datasets of kidney stones, cysts and tumors [50] . Both deep learning techniques (VGG16, Inceptionv3 and Resnet50) and Visual Transformer variants (EANet, CCT and Swin transformer algorithms) can be applied to differentiate KSD from renal cysts and tumors with 99.30 % accuracy achieved by Swin transformer-based model [50] . Caglayan et al [51] have examined the efficacy of a deep learning model for identifying kidney stones in unenhanced CT images in various planes based on stone size.…”

Kidney Diseases Classification using Hybrid Transfer-Learning DenseNet201-Based and Random Forest Classifier

“…Md Nazmul et.al. [13] worked on developing an AI-based system for diagnosing kidney disease and contributed to the AI community's research efforts. This study focuses on the three main categories of renal diseases: cysts, tumor, and kidney stones.…”

“…The suggested autodetection methodology for kidney disease diagnostics used in this study can aid in creating a digital twin of renal function in the context of pathology. This study has bypassed the other studies' performance and contributes greatly to the medical field, specifically for kidney disease detection [13].…”

A transparent artificial intelligence framework to assess lung disease in pulmonary hypertension

“…Shad et al 21 state that studies of explainability, uncertainty and bias should be core components of any clinical AI tool studies. Even though there are studies using explainability techniques to increase the transparency of their AI tools they lack generalized approaches as they mainly use local explainable techniques like salience maps, GradCam, or feature engineering approaches 22 – 24 .…”