Automated and Explainable Kidney Abnormality Detection from CT Images Using CNN LSTM Architecture

In this video, we present “Automated and Explainable Kidney Abnormality Detection from CT Images Using a CNN–LSTM Architecture”—a smart medical imaging pipeline that detects kidney abnormalities from CT scans while also providing explainable visual evidence to improve trust and interpretability.

We explain how CT image slices are preprocessed (noise reduction, normalization, resizing, and ROI handling) and then passed into a CNN to extract strong spatial features such as texture, shape, and lesion patterns. To capture sequential context across multiple CT slices, an LSTM is used to model slice-to-slice dependencies, improving detection of abnormalities like cysts, stones, masses, and structural irregularities.

A key focus of this work is Explainable AI (XAI). We demonstrate how methods like Grad-CAM / saliency maps highlight the regions that influence the model’s decision, helping clinicians and researchers understand why a prediction was made—not just what the prediction is.

✅ What you’ll learn in this video:

* End-to-end workflow for kidney CT abnormality detection
* CNN feature extraction + LSTM sequence learning
* Dataset preparation, augmentation, and training steps
* Performance evaluation (Accuracy, Precision, Recall, F1-score, ROC-AUC)
* Explainability using heatmaps (Grad-CAM)

Applications in clinical decision support and healthcare research

⚠️ Disclaimer: This project is for educational/research purposes and is not a substitute for professional medical diagnosis.

#KidneyCT #MedicalImaging #DeepLearning #CNN #LSTM #ExplainableAI #GradCAM #HealthcareAI #ComputerVision #Radiology #AIinMedicine #ResearchProject

Видео Automated and Explainable Kidney Abnormality Detection from CT Images Using CNN LSTM Architecture канала Jack Sparrow Publishers