Transforming medical diagnostics: federated learning and big data-driven deep learning for precise white blood cell classification

The rise of big data has transformed how complex problems in fields like medicine and biology are solved. In the medical domain, analyzing White Blood Cells (WBCs) is essential for diagnosing diseases and assessing the immune system. While automated tools like cell counters can quickly generate results, manual blood smear analysis remains critical for accuracy and patient monitoring. Unfortunately, this manual process is slow, labour-intensive, and error-prone, making it difficult to manage large-scale data efficiently. This study combines the strengths of Federated Learning and Big Data to tackle these issues. The authors propose a new approach for classifying WBCs by leveraging Federated Learning (FL) for privacy-preserving, distributed training on large datasets while utilizing Apache Spark to manage and process the data. Additionally, advanced deep learning models, such as ResNet50, VGG19, and U-Net, enhance WBC classification accuracy by creating five RDDs and training each of three models on each RDD. RDD has its valuation accuracy. This study addresses the dual challenges of scalability and privacy. Also, distributed data across five Nodes of Resilient Distributed Datasets (RDDs) showed that both VGG19 and ResNet50 achieved higher accuracy than U-Net, while training each deep-learning model to improve diagnostic accuracy by integrating Federated Learning with Big Data frameworks and recent deep-learning techniques. This innovative technique highlights the potential of combining these technologies to advance healthcare and biomedical research.