Alzheimer's disease (AD) remains a formidable challenge in modern healthcare, necessitating innovative approaches for its early detection and intervention. This study aimed to enhance the identification of individuals with mild cognitive impairment (MCI) at risk of developing AD. Leveraging advances in computational power and the extensive availability of healthcare data, we explored the potential of deep learning models for early prediction using medical claims data. We employed a bidirectional gated recurrent unit (BiGRU) deep learning model for predictive modeling of MCI progression across various prediction intervals, extending up to five years post-initial MCI diagnosis. The performance of the BiGRU model was rigorously compared with several machine-learning model baselines to evaluate its efficacy. Using a robust cross-validation methodology, the BiGRU emerged as the top-performing model, achieving an Area Under the Receiver Operating Characteristic Curve (AUC-ROC) of 0.833 (95% CI: 0.822, 0.843), an Area Under the Precision-Recall Curve (AUC-PR) of 0.856 (95% CI: 0.845, 0.867), and an F1-Score of 0.71 (95% CI: 0.694, 0.724) for a five-year prediction interval. The results indicate that BiGRU, utilizing longitudinal claims data, reliably predicts MCI-to-AD progression over a lengthy interval following the initial MCI diagnosis, offering clinicians a valuable tool for targeted risk identification and stratification.

Objective An approach for analysis of hepatitis C virus (HCV) quasispecies using Hiseq high-throughput sequencing (hereinafter referred to as Hiseq sequencing) technique was developed and then applied to investigate a possible case of HCV needle sharing transmission. Methods One case of HCV antibody seroconversion (P1) was found in a methadone clinic on January 15, 2015. Four HCV antibody positive injecting drug users (IDUs), P2 to P5, suspected to be involved in needle sharing transmission with P1 during the period (after March 24, 2014) that P1 may be infected with HCV were investigated, and another 28 HCV antibody positive IDUs were selected as controls (C1 to C28). These controls came from the same methadone clinic or lived in the same town with P1. The RNAs were extracted from the plasma specimens and then reverse-transcribed into cDNA. After HCV subtyping, Hiseq sequencing was performed to detect and sequence the HCV quasispecies (263 bp) in the specimens with the same subtype as P1. The frequency of quasispecies was counted and ranked. Intrapersonal and interpersonal genetic distance and phylogenetic tree were calculated. Results The HCV subtype of specimen P1 was 3b. All the other specimens with the same subtype were P2, C7, C12, C14, C15, C16, C19, C20 and C28. Hiseq sequencing was successfully performed in 9 out of these 10 specimens, and 249 753 to 1 086 333 (average 869 608) cleaned sequences representing 3 to 172 (average 48) unique HCV quasispecies were obtained. The medians (P50) of intrapersonal genetic diversities from the 9 specimens were 0.4% to 12.3%. The P50 (P25, P75) of genetic diversities between P1 and the other 8 specimens were 19.0% (18.4%, 19.8%), 10.4%(2.8%, 18.3%), 19.6% (17.8%, 21.4%),24.9% (23.8%, 26.1%), 19.8% (18.7%, 20.7%), 20.1% (18.9%, 21.2%), 20.6% (20.0%, 21.1%), 23.6% (22.4%, 24.8%). There were no significant difference between the genetic diversities of P1 and P2 and those of P1 and other 7 specimens (H=9.40, P=0.100). The genetic diversities between few HCV quasispecies from P1 and few ones from C7 were 0. Phylogenetic tree analysis indicated that there was no HCV transmission relationship between P1 and P2, but there was HCV transmission relationship between P1 and C7. Conclusion With the feature of high-throughput, easier operation and lower cost, Hiseq sequencing technique has high practical value in tracing HCV transmission at the quasispecies level.

Objective An approach for analysis of hepatitis C virus (HCV) quasispecies using Hiseq high-throughput sequencing (hereinafter referred to as Hiseq sequencing) technique was developed and then applied to investigate a possible case of HCV needle sharing transmission. Methods One case of HCV antibody seroconversion (P1) was found in a methadone clinic on January 15, 2015. Four HCV antibody positive injecting drug users (IDUs), P2 to P5, suspected to be involved in needle sharing transmission with P1 during the period (after March 24, 2014) that P1 may be infected with HCV were investigated, and another 28 HCV antibody positive IDUs were selected as controls (C1 to C28). These controls came from the same methadone clinic or lived in the same town with P1. The RNAs were extracted from the plasma specimens and then reverse-transcribed into cDNA. After HCV subtyping, Hiseq sequencing was performed to detect and sequence the HCV quasispecies (263 bp) in the specimens with the same subtype as P1. The frequency of quasispecies was counted and ranked. Intrapersonal and interpersonal genetic distance and phylogenetic tree were calculated. Results The HCV subtype of specimen P1 was 3b. All the other specimens with the same subtype were P2, C7, C12, C14, C15, C16, C19, C20 and C28. Hiseq sequencing was successfully performed in 9 out of these 10 specimens, and 249 753 to 1 086 333 (average 869 608) cleaned sequences representing 3 to 172 (average 48) unique HCV quasispecies were obtained. The medians (P50) of intrapersonal genetic diversities from the 9 specimens were 0.4% to 12.3%. The P50 (P25, P75) of genetic diversities between P1 and the other 8 specimens were 19.0% (18.4%, 19.8%), 10.4%(2.8%, 18.3%), 19.6% (17.8%, 21.4%),24.9% (23.8%, 26.1%), 19.8% (18.7%, 20.7%), 20.1% (18.9%, 21.2%), 20.6% (20.0%, 21.1%), 23.6% (22.4%, 24.8%). There were no significant difference between the genetic diversities of P1 and P2 and those of P1 and other 7 specimens (H=9.40, P=0.100). The genetic diversities between few HCV quasispecies from P1 and few ones from C7 were 0. Phylogenetic tree analysis indicated that there was no HCV transmission relationship between P1 and P2, but there was HCV transmission relationship between P1 and C7. Conclusion With the feature of high-throughput, easier operation and lower cost, Hiseq sequencing technique has high practical value in tracing HCV transmission at the quasispecies level.

Objective To assess the clinic application of a new-style automated immunoassay system HISCL ( high sensitivity chemiluminescent enzyme immunoassay ) with high sensitive chemiluminescence substrate CDP-Star?, bind-free separate technology and filter conversion technology.Methods The performance verification test evaluated HISCL′s specification including reproducibility, functional sensitivity, linearity, accuracy, and sensitivity for HBsAg seroconversion.To evaluate the specificity , 1 007 HBsAg-negative specimens , 82 potentially interfering specimens were from conventional specimens in West China Hospital , Sichuan University between October and December of 2014.At the same time, with these results to determine the rate of the specimens that their results are in negative grey zone.259 HBsAg-positive specimens and 27 weakly-positive specimens were tested for the comparison between the HISCL and ECLIA ( electrochemiluminescence immunoassay ) HBsAg quantification.A linear regression model , Pearson′s correlation and Bland-Altman analysis were used as statistical methods.Results The between day and within-lot variable coefficient of two level samples are 1.55%, 2.02%, 0.34%, 1.34%separately.The functional sensitivity of the HISCL HBsAg assay reaches 0.007 IU /ml.There is a good linearity (y=3.262x+0.082,r=0.994; y=2 303.608x-33.006,r=0.999) within range of 0-2 300 IU/ml.HISCL obtain the accurate results for the CAP control material , the agreement rate is 100%.The sensitivity for the detection of HBsAg seroconversion is high.The specificity is 99.91%.The negative grey zone rate is 1.66%.The results of these two methods have good correlation and uniformity , r=0.995, LOA:-0.3 -0.19 log10 IU/ml.Conclusions HISCL HBsAg detection system shows good detection performance.Be of the function of both qualitative and quantitative detection.And the negative “grey zone”specimen rate is low.It is wholly capable of wide linearity and quick assays for quantifying serum HBsAg levels.HISCL could contribute to improve HBsAg quantitative detection process for clinical laboratory.