A Deep Learning Model for Classification of Parotid Neoplasms Based on Multimodal Magnetic Resonance Image Sequences

Abstract

Objective: To design a deep learning model based on multimodal magnetic resonance image (MRI) sequences for automatic parotid neoplasm classification, and to improve the diagnostic decision-making in clinical settings.Methods: First, multimodal MRI sequences were collected from 266 patients with parotid neoplasms, and an artificial intelligence (AI)-based deep learning model was designed from scratch, combining the image classification network of Resnet and the Transformer network of Natural language processing. Second, the effectiveness of the deep learning model was improved through the multi-modality fusion of MRI sequences, and the fusion strategy of various MRI sequences was optimized. In addition, we compared the effectiveness of the model in the parotid neoplasm classification with experienced radiologists.Results: The deep learning model delivered reliable outcomes in differentiating benign and malignant parotid neoplasms. The model, which was trained by the fusion of T2-weighted, postcontrast T1-weighted, and diffusion-weighted imaging (b = 1000 s/mm 2 ), produced the best result, with an accuracy score of 0.85, an area under the receiver operator characteristic (ROC) curve of 0.96, a sensitivity score of 0.90, and a specificity score of 0.84. In addition, the multi-modal paradigm exhibited reliable outcomes in diagnosing the pleomorphic adenoma and the Warthin tumor, but not in the identification of the basal cell adenoma.Conclusion: An accurate and efficient AI based classification model was produced to classify parotid neoplasms, resulting from the fusion of multimodal MRI sequences. The effectiveness certainly outperformed the model with single MRI images or single MRI sequences as input, and potentially, experienced radiologists.

Construction and Application of a Nomogram for Predicting Benign and Malignant Parotid Tumors

“…There is still a lack of uniform standards for the preoperative differential diagnosis between benign and malignant parotid tumors. Given the wide diversity of pathologic types of parotid tumors and similarity between their radiographic features, it is difficult to make an accurate diagnosis of benign and malignant parotid tumors based on radiographic features alone 3–5 . Clinicians must rely on their experience when making clinical judgment about parotid tumors, which usually results in a high level of subjectivity and uncertainty, and a greater possibility of misdiagnosis 9 .…”

“…Nomograms offer a reliable statistical prognostic tool by incorporating a variety of risk factors for the differential diagnosis of tumors. 3,5 A visual predictive model can be built based on a nomogram for an individualized, accurate, and evidence-based risk estimate. Recently reported nomograms that have been built based on CT and ultrasound radiomics not only clearly diagnose the types of parotid tumors, but also effectively differentiate malignant from benign parotid tumors.…”

“…There is an urgent need for a more accurate preoperative approach to predict benign and malignant parotid tumors. Nomograms offer a reliable statistical prognostic tool by incorporating a variety of risk factors for the differential diagnosis of tumors 3,5 . A visual predictive model can be built based on a nomogram for an individualized, accurate, and evidence-based risk estimate.…”

Radiomics for Discriminating Benign and Malignant Salivary Gland Tumors; Which Radiomic Feature Categories and MRI Sequences Should Be Used?

“…In radiomics, the discrimination of SGTs have mostly reported results from T1WI, T2WI and DWI [ 18 , 19 , 20 , 21 ]. Only three machine learning study [ 58 , 59 , 60 ], using deep learning instead of radiomics, evaluated CE-T1WI images of the parotid gland. In agreement with our study, one of them found that CE-T1WI did not improve the classification of SGTs [ 59 ].…”

Deep Learning Techniques and Imaging in Otorhinolaryngology—A State-of-the-Art Review

“…Specialized DL algorithms have been developed to assist in differential diagnosis between benign and malignant parotid gland tumors in contrast-enhanced CT images [30], and ultrasonography [31]. MRI remains the gold standard in the diagnosis of salivary gland diseases, where DL models intend to automatically classify salivary gland tumors with very high accuracy [32,33].…”