Radiomics Analysis of DTI Data to Assess Vision Outcome After Intravenous Methylprednisolone Therapy in Neuromyelitis Optic Neuritis

Abstract

Background Neuromyelitis optica‐optic neuritis (NMO‐ON) patients are routinely treated with intravenous methylprednisolone (IVMP). For the patients nonresponsive to IVMP, more effective but aggressive therapy of plasma exchange (PE) should be employed instead of IVMP in the first line. Purpose To assess the visual outcomes of NMO‐ON patients after IVMP by radiomics analysis of whole brain diffusion tensor imaging (DTI) data. Study Type Retrospective. Population In all, 57 NMO‐ON patients receiving IVMP therapy for 3 days. Field Strength/Sequence 3.0T; DTI images acquired by a single‐shot echo planar image sequence; T1 images acquired by 3D fast spoiled gradient echo (3D‐FSPGR) MRI. Assessment In all, 200 DTI measures were extracted from the DTI data and employed as features to construct a radiomics assessment model for visual outcomes of NMO‐ON patients after IVMP. The assessment performance was evaluated by area under the receiver operating characteristic curve (AUC), classification accuracy (ACC), sensitivity, specificity, and positive and negative predicted values (PPV and NPV). The selected DTI measures would reveal the white matter impairments related to visual recovery of NMO‐ON patients. Statistical Tests The relationship between the selected DTI measures and the clinical visual characteristics were investigated by Pearson correlation, Spearman's rank correlation, and one‐way analysis of variance analysis. Results The radiomics model obtained an ACC of 73.68% (P = 0.002), AUC of 0.7931, sensitivity of 0.6207, specificity of 0.8571, PPV of 0.8182, and NPV of 0.6857 in assessing visual outcomes of the NMO‐ON patients after IVMP treatment. The selected DTI measures revealed white matter impairments related to the visual outcomes in the white matter tracts of vision‐relevant regions, motor‐related regions, and corpus callosum. The white matter impairments were found significantly correlated with the disease duration and the length of lesions in the optic nerve. Data Conclusion Radiomics analysis of DTI data has great potential in assessing visual outcomes of NMO‐ON patients after IVMP therapy. Level of Evidence: 2 Technical Efficacy: Stage 4 J. Magn. Reson. Imaging 2019;49:1365–1373.

Radiomic analysis for pretreatment prediction of response to neoadjuvant chemotherapy in locally advanced cervical cancer: A multicentre study

“…Radiomics, as a newly-proposed method, aims to evaluate tumour heterogeneity by extracting high throughput features from medical images that reflect the underlying pathophysiology [[18], [19], [20]]. These radiomic features capture distinct phenotypic differences and may have prognostic and predictive value across different diseases [[21], [22], [23], [24]]. Emerging studies have demonstrated the excellent performance of radiomic models for predicting the pathological complete response (pCR) to NACT in rectal cancer [25,26].…”

Radiomics: from qualitative to quantitative imaging

“…The main requirements for a radiomics study are the presence of a radiologic phenotype which allows for the clustering of patients based on differences within that phenotype or some correlation to the underlying biology, and the availability of imaging and clinical data. While not nearly as prevalent, 127 this has meant that nononcological diseases which require medical imaging as part of the standard of care have also been the subject of radiomics analysis, such as in the fields of neurology, 35 ophthalmology, 128 and dentistry. 129 Limitations of radiomics and future directions towards precision medicine While radiomics facilitates new possibilities in the field of personalised medicine, some challenges remain.…”

Radiomics at a Glance: A Few Lessons Learned from Learning Approaches

“…Similar to fundus photography, a CNN ensemble has been developed to automatically segment and quantify the OCT images, improving prognosis and management of macular diseases [ 62 ]. Radiomic applications have recently appeared too [ 87 , 88 ].…”