Abstract
Objective: We aimed to differentiate granulosa cell tumors (GCT) from other ovarian sex-cord tumors (OSCs) based on feature analysis of the tumor body on MR imaging.
Methods: We retrospectively enrolled 27 patients with pathologically proven sex-cord tumours (14 GSTs, 8 fibromas, 4 fibrothecomas, and 1 sclerosing stromal tumour) from our institution. All MRI examinations were performed at least one month prior to surgery. MR image features were recorded by two radiologists with consensus readings. Histogram analysis was performed using FeAture Explorer software. The differences in histogram parameters between GCT (38.1 ± 14.6 years) and OSC (43.7 ± 18.0 years) groups were compared. Fourteen randomly selected cellular-type myomas who also underwent MRI in our hospital were considered as the control group. The intra-operator consistency of ADC value was evaluated across measurements twice.
Results: The repeatability of conventional ADC measurements on the tumor body was good. The values of ADC-mean, ADC-min, and ADC-max significantly differed across three groups (p < 0.001). The histogram variance on DWI, histogram percentage on T2WI, and ADC min showed the best discriminative performance in determining GCTs from other OSCs with an area under the receiver operator curve (AUC) of 0.997, 0.882, and 0.795, respectively. The histogram variance on DWI yielded a sensitivity of 92.3%, a specificity of 100%, and an accuracy of 96.6% in discriminating GSTs from other OSCs.
Conclusion: In the present study, feature analysis of tumor body MR imaging has helped to differentiate GST from OSC with better performance than conventional ADC measurements.
Keywords: Granulosa cell tumor, Magnetic resonance imaging, Diffusion-weighted imaging, Ovarian neoplasm, Tumor body, Sex cord tumors.
Call for Papers in Thematic Issues
Intelligent Information Retrieval for Multispectral and Hyperspectral Imaging
Multispectral and hyperspectral imaging is an interdisciplinary research area, widely focuses on spectral, spatial, and temporal data. It offers a plethora of opportunities to efficiently analyze the vast area of the earth's surface. However, there are numerous factors such as dimensionality and size of the hyperspectral data, lesser training samples, ...read more
Machine and Deep Learning in Medical Image Diagnosis
Medical Image Processing regards a set of methodologies that have been developed over recent years with the purpose of improving medical image quality, improving medical data visualization, understanding, and assisting medical diagnosis, and so on. The application of machine and deep learning methods in sensing and imaging can potentially have ...read more
Quantum Machine Learning for Medical Data and Imaging
Quantum computing has promised a significant speedup in certain computationally intensive tasks that are intractable on classical computers. Researchers in quantum machine learning, such as those working in computer vision, image processing, biomedical analysis, and related topics, may play an important role in comprehending and working on complicated medical data, ...read more
6