Bamaqa Amna, Labeeb N S, El-Gendy Eman M, Ibrahim Hani M, Farsi Mohamed, Balaha Hossam Magdy, Badawy Mahmoud, Elhosseini Mostafa A
Department of Computer Science and Information, Applied College, Taibah University, Madinah 42353, Saudi Arabia.
Mathematics Department, Faculty of Science, Helwan University, Cairo 11795, Egypt.
Bioengineering (Basel). 2025 Jun 7;12(6):623. doi: 10.3390/bioengineering12060623.
Myeloproliferative neoplasms, particularly the Philadelphia chromosome-negative (Ph-negative) subtypes such as essential thrombocythemia, polycythemia vera, and primary myelofibrosis, present diagnostic challenges due to overlapping morphological features and clinical heterogeneity. Traditional diagnostic approaches, including imaging and histopathological analysis, are often limited by interobserver variability, delayed diagnosis, and subjective interpretations. To address these limitations, we propose a novel framework that integrates handcrafted and automatic feature extraction techniques for improved classification of Ph-negative myeloproliferative neoplasms. Handcrafted features capture interpretable morphological and textural characteristics. In contrast, automatic features utilize deep learning models to identify complex patterns in histopathological images. The extracted features were used to train machine learning models, with hyperparameter optimization performed using Optuna. Our framework achieved high performance across multiple metrics, including precision, recall, F1 score, accuracy, specificity, and weighted average. The concatenated probabilities, which combine both feature types, demonstrated the highest mean weighted average of 0.9969, surpassing the individual performances of handcrafted (0.9765) and embedded features (0.9686). Statistical analysis confirmed the robustness and reliability of the results. However, challenges remain in assuming normal distributions for certain feature types. This study highlights the potential of combining domain-specific knowledge with data-driven approaches to enhance diagnostic accuracy and support clinical decision-making.
骨髓增殖性肿瘤，尤其是费城染色体阴性（Ph阴性）亚型，如原发性血小板增多症、真性红细胞增多症和原发性骨髓纤维化，由于形态学特征重叠和临床异质性，给诊断带来了挑战。传统的诊断方法，包括影像学和组织病理学分析，常常受到观察者间差异、诊断延迟和主观解释的限制。为了解决这些局限性，我们提出了一个新颖的框架，该框架整合了手工制作和自动特征提取技术，以改进对Ph阴性骨髓增殖性肿瘤的分类。手工制作的特征捕捉可解释的形态学和纹理特征。相比之下，自动特征利用深度学习模型来识别组织病理学图像中的复杂模式。提取的特征用于训练机器学习模型，并使用Optuna进行超参数优化。我们的框架在多个指标上都取得了高性能，包括精确率、召回率、F1分数、准确率、特异性和加权平均值。结合了两种特征类型的联合概率显示出最高的平均加权平均值，为0.9969，超过了手工制作特征（0.9765）和嵌入特征（0.9686）的个体性能。统计分析证实了结果的稳健性和可靠性。然而，在假设某些特征类型呈正态分布方面仍然存在挑战。这项研究突出了将特定领域知识与数据驱动方法相结合以提高诊断准确性并支持临床决策的潜力。
