Prion, 2019. 13(1), 141-150, DOI: https://doi.org/10.1080/19336896.2019.1639482

Sensitivity and specificity evaluation of multiple neurodegenerative proteins for Creutzfeldt-Jakob disease diagnosis using a deep-learning approach

Sol Moe Lee, Jae Wook Hyeon;Soo-Jin Kim;Heebal Kim;Ran Noh;Seonghan Kim;Yeong Seon Lee;Su Yeon Kim

Abstract

The diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD) can only be confirmed by abnormalprotease-resistant prion protein accumulation in post-mortem brain tissue. The relationshipsbetween sCJD and cerebrospinal fluid (CSF) proteins such as 14–3-3, tau, and α-synuclein(a-syn) have been investigated for their potential value in pre-mortem diagnosis. Recently, deeplearning(DL) methods have attracted attention in neurodegenerative disease research. Weestablished DL-aided pre-mortem diagnostic methods for CJD using multiple CSF biomarkers toimprove their discriminatory sensitivity and specificity. Enzyme-linked immunosorbent assayswere performed on phospho-tau (p-tau), total-tau (t-tau), a-syn, and β-amyloid (1–42), andwestern blot analysis was performed for 14–3-3 protein from CSF samples of 49 sCJD and 256non-CJD Korean patients, respectively. The deep neural network structure comprised one input,five hidden, and one output layers, with 20, 40, 30, 20 and 12 hidden unit numbers per hiddenlayer, respectively. The best performing DL model demonstrated 90.38% accuracy, 83.33% sensitivity,and 92.5% specificity for the three-protein combination of t-tau, p-tau, and a-syn, and allother patients in a separate CSF set (n = 15) with other neuronal diseases were correctly predictedto not have CJD. Thus, DL-aided pre-mortem diagnosis may provide a suitable tool for discriminatingCJD patients from non-CJD patients.

• DOI: https://doi.org/10.1080/19336896.2019.1639482
• ISBN or ISSN: 1933-6896

• 본 연구는 질병관리본부 연구개발과제연구비를 지원받아 수행되었습니다.
• This research was supported by a fund by Research of Korea Centers for Disease Control and Prevention.