A novel transfer learning model for predictive analytics using incomplete multimodality data. (16th June 2021)

• Record Type: Journal Article
• Authors: Liu, Xiaonan; Chen, Kewei; Weidman, David; Wu, Teresa; Lure, Fleming; Li, Jing
• Is Part Of: IISE transactions. Volume 53:Number 9(2021)
• Page Start: 1010
• Page End: 1022
• Publication Date: 2021-06-16
• Subjects: Incomplete multimodality data -- transfer learning -- predictive analytics -- health care; Industrial engineering -- Periodicals; Systems engineering -- Periodicals; Industrial engineering; Systems engineering; Electronic journals; Periodicals; 670.285
• DOI: 10.1080/24725854.2020.1798569
• Languages: English
• ISSNs: 2472-5854

Abstract

Abstract: Multimodality datasets are becoming increasingly common in various domains to provide complementary information for predictive analytics. One significant challenge in fusing multimodality data is that the multiple modalities are not universally available for all samples due to cost and accessibility constraints. This situation results in a unique data structure called an Incomplete Multimodality Dataset. We propose a novel Incomplete-Multimodality Transfer Learning (IMTL) model that builds a predictive model for each sub-cohort of samples with the same missing modality pattern, and meanwhile couples the model estimation processes for different sub-cohorts to allow for transfer learning. We develop an Expectation-Maximization (EM) algorithm to estimate the parameters of IMTL and further extend it to a collaborative learning paradigm that is specifically valuable for patient privacy preservation in health care applications. We prove two advantageous properties of IMTL: the ability for out-of-sample prediction and a theoretical guarantee for a larger Fisher information compared with models without transfer learning. IMTL is applied to diagnosis and prognosis of Alzheimer's disease at an early stage called Mild Cognitive Impairment using incomplete multimodality imaging data. IMTL achieves higher accuracy than competing methods without transfer learning.