# Spectral-aware CNN with learnable biorthogonal units and depthwise convolutions for multi-class blood cell classification

## TL;DR

This paper introduces a new deep learning model for classifying blood cells with high accuracy using advanced convolution techniques.

## Contribution

The novel model uses learnable spectral biorthogonal units and depthwise separable convolutions to improve classification efficiency and accuracy.

## Key findings

- The model achieves 99.18% overall classification accuracy on a dataset of 17,092 images.
- The model demonstrates superior generalization across all blood cell classes without overfitting.
- The use of LSBDUs and depthwise separable convolutions reduces computational overhead while preserving features.

## Abstract

For effective and early diagnosis of diseases such as leukemia and anemia, accurate classification and interpretation of peripheral blood cells are critical. A novel hybrid deep learning model is proposed in this study for multi-class blood cell classification, called Spectral-Aware CNN with Learnable Spectral Biorthogonal Downsampling Units (LSBDUs) and Depthwise Separable Convolutions. The model replaces conventional pooling layers with wavelet-inspired LSBDUs for improved feature retention. This results in reduced computational overhead through efficient separable convolutions. The research used a balanced dataset of 17,092 images across eight blood cell classes. The techniques, such as stratified data splitting, advanced augmentation, and label smoothing, are included in the training pipeline for improving generalizability. As a result, the model achieves 99.18 % of overall classification accuracy with superior class-wise performance.•Replaces pooling layers with spectral-aware LSBDU blocks for better feature preservation.•Integrates Depthwise Separable Convolutions to reduce parameter count and training cost.•Demonstrates superior generalization across all classes without overfitting.

Replaces pooling layers with spectral-aware LSBDU blocks for better feature preservation.

Integrates Depthwise Separable Convolutions to reduce parameter count and training cost.

Demonstrates superior generalization across all classes without overfitting.

Image, graphical abstract