Machine learning algorithms to automate differentiating cardiac amyloidosis from hypertrophic cardiomyopathy.

Cardiac amyloidosis has a poor prognosis, and high mortality and is often misdiagnosed as hypertrophic cardiomyopathy, leading to delayed diagnosis. Machine learning combined with speckle tracking echocardiography was proposed to automate differentiating two conditions. A total of 74 patients with pathologically confirmed monoclonal immunoglobulin light chain cardiac amyloidosis and 64 patients with hypertrophic cardiomyopathy were enrolled from June 2015 to November 2018. Machine learning models utilizing traditional and advanced algorithms were established and determined the most significant predictors. The performance was evaluated by the receiver operating characteristic curve (ROC) and the area under the curve (AUC). With clinical and echocardiography data, all models showed great discriminative performance (AUC > 0.9). Compared with logistic regression (AUC 0.91), machine learning such as support vector machine (AUC 0.95, p = 0.477), random forest (AUC 0.97, p = 0.301) and gradient boosting machine (AUC 0.98, p = 0.230) demonstrated similar capability to distinguish cardiac amyloidosis and hypertrophic cardiomyopathy. With speckle tracking echocardiography, the predictive performance of the voting model was similar to that of LightGBM (AUC was 0.86 for both), while the AUC of XGBoost was slightly lower (AUC 0.84). In fivefold cross-validation, the voting model was more robust globally and superior to the single model in some test sets. Data-driven machine learning had shown admirable performance in differentiating two conditions and could automatically integrate abundant variables to identify the most discriminating predictors without making preassumptions. In the era of big data, automated machine learning will help to identify patients with cardiac amyloidosis and timely and effectively intervene, thus improving the outcome.

A case report-application of pericardial effusion cytology and next-generation sequencing technology: quick and secure diagnosis of primary effusion lymphoma.

Background: Primary effusion lymphoma (PEL) is an uncommon subtype of non-Hodgkin lymphoma (NHL) that usually involves the pleura, pericardium, and peritoneum without an obvious tumour mass, with multiple plasma effusions as its main clinical feature. We report a case of a massive pericardial effusion in an elderly male with a final diagnosis of PEL. Case summary: A 70-year-old male patient was admitted to hospital with symptoms of chest tightness, shortness of breath, fatigue, loss of appetite, and cough with phlegm after a pericardial effusion had been found for 5 months. The next-generation sequencing of pericardial effusion found human herpesvirus type 8 (HHV-8) infection, and further cytomorphological and immunohistochemical examination were done. According to the patient's HHV-8 infection, the pathological features of heterogeneous B cells with plasmablastic differentiation and the immunohistochemical characteristics of PEL, the final diagnosis was made as human immunodeficiency virus-negative PEL. Discussion: The diversity and non-specificity of PEL symptoms, as well as its rarity, make it difficult to diagnose. In this case, we used the next-generation sequencing technology to screen the pathogen of the patient's pericardial effusion and carried out morphological and immunohistochemical examination of the cells in the pericardial effusion, which provided a clinically operable diagnosis for an uncommon disease, enabling us to make a clear diagnosis faster and start treatment in time.

miR-375 Induced the Formation and Transgenerational Inheritance of Fatty Liver in Poultry by Targeting MAP3K1.

The liver of poultry is the primary site of lipid synthesis. The excessive production of lipids accumulates in liver tissues causing lipid metabolism disorders, which result in fatty liver disease and have a transgenerational effect of acquired phenotypes. However, its specific mechanisms have not yet been fully understood. In this study, the differentially expressed miR-375 as well as its target gene MAP3K1 (mitogen-activated protein kinase kinase kinase 1) were screened out by interaction network analysis of microRNA sequencing results and transcriptome profiling in the fatty liver group of the F0-F3 generation (p < 0.05 or p < 0.01). Furthermore, the results showed that the number of lipid droplets and triglyceride content were significantly decreased after upregulation of miR-375 in primary hepatocyte culture in vitro (p < 0.05 or p < 0.01). The MAP3K1 knockdown group exhibited the opposite trends (p < 0.05 or p < 0.01). P53, Bcl-x, PMP22, and CDKN2C related to cell proliferation were significantly upregulated or downregulated after knocking down MAP3K1 (p < 0.05). This research uniquely revealed that silencing miR-375 inhibits lipid biosynthesis and promotes cell proliferation, which may be due to the partial regulation of the expression level of MAP3K1, thereby further participating in the transgenerational inheritance process of regulating liver lipid metabolism. These results reveal the pathogenesis of fatty liver in noncoding RNA and provide good candidate genes for breeding progress of disease resistance in chickens.

[Survey of HIV drug resistance threshold in Zhejiang province from 2009 to 2011].

OBJECTIVE: To survey the prevalence of drug resistant HIV in Zhejiang province in 2009-2011. METHODS: WHO truncated sequential sampling technique was adopted annually by using 63, 62 and 57 samples of newly diagnosed as HIV positive and aged 16-25 years in Hangzhou, Ningbo and Wenzhou from 2009 to 2011, respectively. RNA was prepared and HIV pol region was amplified by RT-PCR and nested PCR. Pol genetic mutation associated with drug resistance was analyzed. RESULTS: The success rates for sequence acquisition of the survey were 82.5% (52/63), 95.2% (59/62) and 94.7% (54/57) from year 2009 to 2011, respectively, and the main subtype was CRF01_AE (68.5% (37/54)-71.2% (37/52)). A total of 4 surveillance drug-resistance mutation (SDRMs), 2 SDRMs and 2 SDRMs were found by analyzing the 47 sequences each year, sampled from year 2009 to 2010, respectively, indicating that the prevalence of drug resistant HIV stains was moderate in 2009, and low for the next two years (2010-2011). A total of 8 individuals with drug resistant HIV stains found in this study were all infected by sexual transmission, especially in homosexual transmission (6 cases), and the main subtype was CRF01_AE (7 cases). SDRMs for protease inhibitor (PI), nucleotide HIV-reverse transcriptase inhibitor (NRTI) and non-NRTI (NNRTI) (L90M, T215S and Y188L) were all found in one case. CONCLUSION: The prevalence of drug resistant HIV stains in major areas with AIDS epidemic in Zhejiang province was low in 2009-2011.