In Heliyon

BACKGROUND AND OBJECTIVE : A hyperinflammatory environment is thought to be the distinctive characteristic of COVID-19 infection and an important mediator of morbidity. This study aimed to determine the effect of other immunological parameter levels, especially ferritin, as a predictor of COVID-19 mortality via decision-trees analysis.

MATERIAL AND METHOD : This is a retrospective study evaluating a total of 2568 patients who died (n = 232) and recovered (n = 2336) from COVID-19 in August and December 2021. Immunological laboratory data were compared between two groups that died and recovered from patients with COVID-19. In addition, decision trees from machine learning models were used to evaluate the performance of immunological parameters in the mortality of the COVID-19 disease.

RESULTS : Non-surviving from COVID-19 had 1.75 times higher ferritin, 10.7 times higher CRP, 2.4 times higher D-dimer, 1.14 times higher international-normalized-ratio (INR), 1.1 times higher Fibrinogen, 22.9 times higher procalcitonin, 3.35 times higher troponin, 2.77 mm/h times higher erythrocyte-sedimentation-rate (ESR), 1.13sec times longer prothrombin time (PT) when compared surviving patients. In addition, our interpretable decision tree, which was constructed with only the cut-off values of ferritin, INR, and D-dimer, correctly predicted 99.7% of surviving patients and 92.7% of non-surviving patients.

CONCLUSIONS : This study perfectly predicted the mortality of COVID-19 with our interpretable decision tree constructed with INR and D-dimer, especially ferritin. For this reason, we think that it may be important to include ferritin, INR, and D-dimer parameters and their cut-off values in the scoring systems to be planned for COVID-19 mortality.

Huyut Mehmet Tahir, Huyut Zübeyir

2023-Mar

Artificial intelligence, CHAID decision Trees, COVID-19, Coagulation tests, Ferritin, Immunological tests, Machine learning, Mortality risk biomarkers

In Porcine health management

BACKGROUND : Good estrus detection in sows is essential to predict the best moment of insemination. Nowadays, a technological innovation is available that detects the estrus of the sow via connected sensors and cameras. The collected data are subsequently analyzed by an artificial intelligence (AI) system. This study investigated whether such an AI system could support the farmer in optimizing the moment of insemination and reproductive performance. M&M: Three Belgian sow farms (A, B and C) where the AI system was installed, participated in the study. The reproductive cycles (n = 6717) of 1.5 years before and 1.5 years after implementation of the system were included. Parameters included: (1) farrowing rate (FR), (2) percentage of repeat-breeders (RB), (3) farrowing rate after first insemination (FRFI) and (4) number of total born piglets per litter (NTBP). Also, data collected by the system were analyzed to describe the weaning-to-estrus interval (WEI), estrus duration (ED) and the number of inseminations used per estrus. This dataset included 2261 cycles, collected on farms B and C.

RESULTS : In farm A, all parameters significantly improved namely FR + 4.3%, RB - 3.75%, FRFI + 6.2% and NTBP + 1.06 piglets. In farm B, the NTBP significantly decreased with 0.48 piglets, but in this farm the insemination dose was too low (0.8 × 10⁹ spermatozoa per dose). In farm C, only the NTBP significantly increased with 0.45 piglets after the implementation of the system. The WEI as determined by the system varied between 78 and 90 h, being 10-20 h shorter in comparison with the WEI as determined by the farmer. The ED, determined by the system ranged from 48 to 60 h, and was less variable as compared to the ED as assessed by the farmer. The mean number of inseminations per estrus remained similar over time in farm B whereas it decreased over time from approximately 1.6-1.2 in farm C.

CONCLUSION : The AI system can help farmers to improve the reproductive performance, assess estrus characteristics and reduce the number of inseminations per estrus. Results may vary between farms as many other variables such as farm management, genetics and insemination dose also influence reproductive performance.

Verhoeven Steven, Chantziaras Ilias, Bernaerdt Elise, Loicq Michel, Verhoeven Ludo, Maes Dominiek

2023-Mar-15

Artificial intelligence, Estrus detection, Pig production, Reproductive performance

In Parasites & vectors ; h5-index 57.0

BACKGROUND : Apicomplexa consist of numerous pathogenic parasitic protistan genera that invade host cells and reside and replicate within the parasitophorous vacuole (PV). Through this interface, the parasite exchanges nutrients and affects transport and immune modulation. During the intracellular life-cycle, the specialized secretory organelles of the parasite secrete an array of proteins, among which dense granule proteins (GRAs) play a major role in the modification of the PV. Despite this important role of GRAs, a large number of potential GRAs remain unidentified in Apicomplexa.

METHODS : A multi-view attention graph convolutional network (MVA-GCN) prediction model with multiple features was constructed using a combination of machine learning and genomic datasets, and the prediction was performed on selected Neospora caninum protein data. The candidate GRAs were verified by a CRISPR/Cas9 gene editing system, and the complete NcGRA64(a,b) gene knockout strain was constructed and the phenotypes of the mutant were analyzed.

RESULTS : The MVA-GCN prediction model was used to screen N. caninum candidate GRAs, and two novel GRAs (NcGRA64a and NcGRA64b) were verified by gene endogenous tagging. Knockout of complete genes of NcGRA64(a,b) in N. caninum did not affect the parasite's growth and replication in vitro and virulence in vivo.

CONCLUSIONS : Our study showcases the utility of the MVA-GCN deep learning model for mining Apicomplexa GRAs in genomic datasets, and the prediction model also has certain potential in mining other functional proteins of apicomplexan parasites.

Lu Zhenxiao, Hu Hang, Song Yashan, Zhou Siyi, Ayanniyi Olalekan Opeyemi, Xu Qianming, Yue Zhenyu, Yang Congshan

2023-Mar-14

Apicomplexa, Dense granule protein, MVA-GCN, Machine learning, Parasites

In BMC medical genomics

BACKGROUND : Increasing evidence has indicated that ferroptosis engages in the progression of Parkinson's disease (PD). This study aimed to explore the role of ferroptosis-related genes (FRGs), immune infiltration and immune checkpoint genes (ICGs) in the pathogenesis and development of PD.

METHODS : The microarray data of PD patients and healthy controls (HC) from the Gene Expression Omnibus (GEO) database was downloaded. Weighted gene co-expression network analysis (WGCNA) was processed to identify the significant modules related to PD in the GSE18838 dataset. Machine learning algorithms were used to screen the candidate biomarkers based on the intersect between WGCNA, FRGs and differentially expressed genes. Enrichment analysis of GSVA, GSEA, GO, KEGG, and immune infiltration, group comparison of ICGs were also performed. Next, candidate biomarkers were validated in clinical samples by ELISA and receiver operating characteristic curve (ROC) was used to assess diagnose ability.

RESULTS : In this study, FRGs had correlations with ICGs, immune infiltration. Then, plasma levels of LPIN1 in PD was significantly lower than that in healthy controls, while the expression of TNFAIP3 was higher in PD in comparison with HC. ROC curves showed that the area under curve (AUC) of the LPIN1 and TNFAIP3 combination was 0.833 (95% CI: 0.750-0.916). Moreover, each biomarker alone could discriminate the PD from HC (LPIN1: AUC = 0.754, 95% CI: 0.659-0.849; TNFAIP3: AUC = 0.754, 95% CI: 0.660-0.849). For detection of early PD from HC, the model of combination maintained diagnostic accuracy with an AUC of 0.831 (95% CI: 0.734-0.927), LPIN1 also performed well in distinguishing the early PD from HC (AUC = 0.817, 95% CI: 0.717-0.917). However, the diagnostic efficacy was relatively poor in distinguishing the early from middle-advanced PD patients.

CONCLUSION : The combination model composed of LPIN1 and TNFAIP3, and each biomarker may serve as an efficient tool for distinguishing PD from HC.

Xing Na, Dong Ziye, Wu Qiaoli, Zhang Yufeng, Kan Pengcheng, Han Yuan, Cheng Xiuli, Wang Yaru, Zhang Biao

2023-Mar-14

Bioinformatic, ELISA, Ferroptosis, Immune checkpoint gene, Immune infiltration, Parkinson’s disease

ArXiv Preprint

Jun Yu, Jichao Zhu, Wangyuan Zhu, Zhongpeng Cai, Guochen Xie, Renda Li, Gongpeng Zhao

2023-03-16

In BMC musculoskeletal disorders ; h5-index 46.0

Gupta Latika, Najm Aurélie, Kabir Koroush, De Cock Diederik

2023-Mar-14

Artificial intelligence, Digital health, Musculoskeletal, Orthopedics, Rheumatology, Telemedicine, Virtual consultation, eHealth