Rao Divya, Singh Rohit, Koteshwara Prakashini, Vijayananda J
Department of Information and Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, 576104 India.
Department of Otorhinolaryngology, Kasturba Medical College, Manipal Academy of Higher Education, Manipal, 576104 India.
Indian J Otolaryngol Head Neck Surg. 2024 Oct;76(5):4036-4042. doi: 10.1007/s12070-024-04776-8. Epub 2024 Jun 6.
: Laryngeal cancer accounts for a third of all head and neck malignancies, necessitating timely detection for effective treatment and enhanced patient outcomes. Machine learning shows promise in medical diagnostics, but the impact of model complexity on diagnostic efficacy in laryngeal cancer detection can be ambiguous. : In this study, we examine the relationship between model sophistication and diagnostic efficacy by evaluating three approaches: Logistic Regression, a small neural network with 4 layers of neurons and a more complex convolutional neural network with 50 layers and examine their efficacy on laryngeal cancer detection on computed tomography images. : Logistic regression achieved 82.5% accuracy. The 4-Layer NN reached 87.2% accuracy, while ResNet-50, a deep learning architecture, achieved the highest accuracy at 92.6%. Its deep learning capabilities excelled in discerning fine-grained CT image features. : Our study highlights the choices involved in selecting a laryngeal cancer detection model. Logistic regression is interpretable but may struggle with complex patterns. The 4-Layer NN balances complexity and accuracy. ResNet-50 excels in image classification but demands resources. This research advances understanding affect machine learning model complexity could have on learning features of laryngeal tumor features in contrast CT images for purposes of disease prediction.
喉癌占所有头颈恶性肿瘤的三分之一，因此需要及时检测以进行有效治疗并改善患者预后。机器学习在医学诊断中显示出前景，但模型复杂性对喉癌检测诊断效能的影响可能并不明确。在本研究中，我们通过评估三种方法来考察模型复杂度与诊断效能之间的关系：逻辑回归、具有4层神经元的小型神经网络以及具有50层的更复杂的卷积神经网络，并考察它们在计算机断层扫描图像上检测喉癌的效能。逻辑回归的准确率达到82.5%。4层神经网络的准确率达到87.2%，而深度学习架构ResNet - 50的准确率最高，为92.6%。其深度学习能力在辨别细粒度CT图像特征方面表现出色。我们的研究突出了选择喉癌检测模型时涉及的各种考量。逻辑回归具有可解释性，但可能难以处理复杂模式。4层神经网络在复杂性和准确性之间取得了平衡。ResNet - 50在图像分类方面表现出色，但需要资源。这项研究增进了我们对于机器学习模型复杂性在对比CT图像中学习喉肿瘤特征以进行疾病预测方面可能产生的影响的理解。
