Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Adaptive Local Binary Pattern: A Novel Feature Descriptor for Enhanced Analysis of Kidney Abnormalities in CT Scan Images using ensemble based Machine Learning Approach

Tahmim Hossain, Faisal Sayed, Solehin Islam

TL;DR

The paper tackles automated detection of kidney abnormalities in CT scans, addressing clinician shortages by introducing a preprocessing+texture-descriptor pipeline centered on Adaptive Local Binary Pattern ($A ext{-}LBP$) and a soft voting ensemble. It couples cropping, resizing to 224×224, and CLAHE with both LBP and the novel $A ext{-}LBP$ descriptor, and evaluates five classifiers (Random Forest, Decision Tree, Naive Bayes, K-NN, SVM) under a soft voting framework. On a Dhaka-based dataset of 12,446 CT images across cyst, normal, stone, and tumor classes, the approach achieves up to around $0.99$ accuracy, significantly outperforming traditional LBP baselines. The work highlights the practical potential of a CPU-friendly, handcrafted-texture descriptor for reliable, high-precision kidney-abnormality classification and outlines path forward for larger-scale validation and clinical integration. The key novelty is the $A ext{-}LBP$ descriptor, which leverages an adaptive threshold $\beta$ to capture nuanced local textures when combined with CLAHE and ensemble decision fusion.

Abstract

The shortage of nephrologists and the growing public health concern over renal failure have spurred the demand for AI systems capable of autonomously detecting kidney abnormalities. Renal failure, marked by a gradual decline in kidney function, can result from factors like cysts, stones, and tumors. Chronic kidney disease may go unnoticed initially, leading to untreated cases until they reach an advanced stage. The dataset, comprising 12,427 images from multiple hospitals in Dhaka, was categorized into four groups: cyst, tumor, stone, and normal. Our methodology aims to enhance CT scan image quality using Cropping, Resizing, and CALHE techniques, followed by feature extraction with our proposed Adaptive Local Binary Pattern (A-LBP) feature extraction method compared with the state-of-the-art local binary pattern (LBP) method. Our proposed features fed into classifiers such as Random Forest, Decision Tree, Naive Bayes, K-Nearest Neighbor, and SVM. We explored an ensemble model with soft voting to get a more robust model for our task. We got the highest of more than 99% in accuracy using our feature descriptor and ensembling five classifiers (Random Forest, Decision Tree, Naive Bayes, K-Nearest Neighbor, Support Vector Machine) with the soft voting method.

Adaptive Local Binary Pattern: A Novel Feature Descriptor for Enhanced Analysis of Kidney Abnormalities in CT Scan Images using ensemble based Machine Learning Approach

TL;DR

The paper tackles automated detection of kidney abnormalities in CT scans, addressing clinician shortages by introducing a preprocessing+texture-descriptor pipeline centered on Adaptive Local Binary Pattern () and a soft voting ensemble. It couples cropping, resizing to 224×224, and CLAHE with both LBP and the novel descriptor, and evaluates five classifiers (Random Forest, Decision Tree, Naive Bayes, K-NN, SVM) under a soft voting framework. On a Dhaka-based dataset of 12,446 CT images across cyst, normal, stone, and tumor classes, the approach achieves up to around accuracy, significantly outperforming traditional LBP baselines. The work highlights the practical potential of a CPU-friendly, handcrafted-texture descriptor for reliable, high-precision kidney-abnormality classification and outlines path forward for larger-scale validation and clinical integration. The key novelty is the descriptor, which leverages an adaptive threshold to capture nuanced local textures when combined with CLAHE and ensemble decision fusion.

Abstract

The shortage of nephrologists and the growing public health concern over renal failure have spurred the demand for AI systems capable of autonomously detecting kidney abnormalities. Renal failure, marked by a gradual decline in kidney function, can result from factors like cysts, stones, and tumors. Chronic kidney disease may go unnoticed initially, leading to untreated cases until they reach an advanced stage. The dataset, comprising 12,427 images from multiple hospitals in Dhaka, was categorized into four groups: cyst, tumor, stone, and normal. Our methodology aims to enhance CT scan image quality using Cropping, Resizing, and CALHE techniques, followed by feature extraction with our proposed Adaptive Local Binary Pattern (A-LBP) feature extraction method compared with the state-of-the-art local binary pattern (LBP) method. Our proposed features fed into classifiers such as Random Forest, Decision Tree, Naive Bayes, K-Nearest Neighbor, and SVM. We explored an ensemble model with soft voting to get a more robust model for our task. We got the highest of more than 99% in accuracy using our feature descriptor and ensembling five classifiers (Random Forest, Decision Tree, Naive Bayes, K-Nearest Neighbor, Support Vector Machine) with the soft voting method.
Paper Structure (27 sections, 2 equations, 4 figures, 5 tables, 1 algorithm)

This paper contains 27 sections, 2 equations, 4 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Proposed Feature Extraction Method
  4. Dataset
  5. Proposed Framework
  6. Data Preprocessing
  7. Data Cropping & Resizing
  8. Apply Contrast-Limited Adaptive Histogram Equalization (CLAHE) to the Data
  9. Feature Extraction
  10. Local Binary Pattern (LBP)
  11. Our Proposed Feature Extraction Method, Adaptive Local Binary Pattern (A-LBP)
  12. Classifier Selection
  13. Random Forest
  14. Decision Tree
  15. Naive Bayes
  16. ...and 12 more sections

Figures (4)

  • Figure 1: The encoding processes of LBP and A-LBP in a pixel block.
  • Figure 2: Proposed Methodology
  • Figure 3: (a) Raw Normal Kidney Image (b) Cropped Image (c) Resized Image (d) CLAHE Applied Image
  • Figure 4: Four types of kidney image slices after applying Local Binary Pattern (LBP) and our proposed feature descriptor, Adaptive Local Binary Pattern(A-LBP)