1
|
Shi M, Lin J, Wei W, Qin Y, Meng S, Chen X, Li Y, Chen R, Yuan Z, Qin Y, Huang J, Liang B, Liao Y, Ye L, Liang H, Xie Z, Jiang J. Machine learning-based in-hospital mortality prediction of HIV/AIDS patients with Talaromyces marneffei infection in Guangxi, China. PLoS Negl Trop Dis 2022; 16:e0010388. [PMID: 35507586 PMCID: PMC9067679 DOI: 10.1371/journal.pntd.0010388] [Citation(s) in RCA: 10] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/23/2021] [Accepted: 04/02/2022] [Indexed: 12/03/2022] Open
Abstract
Objective Talaromycosis is a serious regional disease endemic in Southeast Asia. In China, Talaromyces marneffei (T. marneffei) infections is mainly concentrated in the southern region, especially in Guangxi, and cause considerable in-hospital mortality in HIV-infected individuals. Currently, the factors that influence in-hospital death of HIV/AIDS patients with T. marneffei infection are not completely clear. Existing machine learning techniques can be used to develop a predictive model to identify relevant prognostic factors to predict death and appears to be essential to reducing in-hospital mortality. Methods We prospectively enrolled HIV/AIDS patients with talaromycosis in the Fourth People’s Hospital of Nanning, Guangxi, from January 2012 to June 2019. Clinical features were selected and used to train four different machine learning models (logistic regression, XGBoost, KNN, and SVM) to predict the treatment outcome of hospitalized patients, and 30% internal validation was used to evaluate the performance of models. Machine learning model performance was assessed according to a range of learning metrics, including area under the receiver operating characteristic curve (AUC). The SHapley Additive exPlanations (SHAP) tool was used to explain the model. Results A total of 1927 HIV/AIDS patients with T. marneffei infection were included. The average in-hospital mortality rate was 13.3% (256/1927) from 2012 to 2019. The most common complications/coinfections were pneumonia (68.9%), followed by oral candida (47.5%), and tuberculosis (40.6%). Deceased patients showed higher CD4/CD8 ratios, aspartate aminotransferase (AST) levels, creatinine levels, urea levels, uric acid (UA) levels, lactate dehydrogenase (LDH) levels, total bilirubin levels, creatine kinase levels, white blood-cell counts (WBC) counts, neutrophil counts, procaicltonin levels and C-reactive protein (CRP) levels and lower CD3+ T-cell count, CD8+ T-cell count, and lymphocyte counts, platelet (PLT), high-density lipoprotein cholesterol (HDL), hemoglobin (Hb) levels than those of surviving patients. The predictive XGBoost model exhibited 0.71 sensitivity, 0.99 specificity, and 0.97 AUC in the training dataset, and our outcome prediction model provided robust discrimination in the testing dataset, showing an AUC of 0.90 with 0.69 sensitivity and 0.96 specificity. The other three models were ruled out due to poor performance. Septic shock and respiratory failure were the most important predictive features, followed by uric acid, urea, platelets, and the AST/ALT ratios. Conclusion The XGBoost machine learning model is a good predictor in the hospitalization outcome of HIV/AIDS patients with T. marneffei infection. The model may have potential application in mortality prediction and high-risk factor identification in the talaromycosis population. Talaromyces marneffei can cause a fatal deeply disseminated fungal infection- talaromycosis. It is widely distributed in Southeast Asia and spreading globally, the disease is insidious and responsible for significant deaths. Clinicians need easy-to-use tools to make decisions on which patients are at a higher risk of dying after infecting T. marneffei. In this study, conducted in Southern China, we have evolved XGBoost machine learning model. 15 clinical indicators and laboratory measures were used to estimate a patient’s risk of dying in the hospital due to the T. marneffei infection. The study showed that the machine learning model has good predictive ability when tested in an internal testing population of patients. We expect that the model could help clinicians assess a patient’s risk of death in just the time of admission to help decide on early treatment timing of high-risk patients who are likely to die.
Collapse
Affiliation(s)
- Minjuan Shi
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
| | - Jianyan Lin
- Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
| | - Wudi Wei
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
| | - Yaqin Qin
- Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
| | - Sirun Meng
- Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
| | - Xiaoyu Chen
- Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
| | - Yueqi Li
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
| | - Rongfeng Chen
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
| | - Zongxiang Yuan
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
| | - Yingmei Qin
- Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
| | - Jiegang Huang
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
| | - Bingyu Liang
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
| | - Yanyan Liao
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
| | - Li Ye
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
- * E-mail: (LY); (HL); (ZX); (JJ)
| | - Hao Liang
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
- * E-mail: (LY); (HL); (ZX); (JJ)
| | - Zhiman Xie
- Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
- * E-mail: (LY); (HL); (ZX); (JJ)
| | - Junjun Jiang
- Guangxi Key Laboratory of AIDS Prevention and Treatment & School of Public Health, Guangxi Medical University, Nanning, Guangxi, China
- Joint Laboratory for Emerging Infectious Diseases in China (Guangxi)-ASEAN, Life Sciences Institute, Guangxi Medical University, Nanning, Guangxi, China
- * E-mail: (LY); (HL); (ZX); (JJ)
| |
Collapse
|
2
|
Su J, Qin Z, Fu H, Luo J, Huang Y, Huang P, Zhang S, Liu T, Lu W, Li W, Jiang T, Wei S, Yang S, Shen Y. Association of prenatal renal ultrasound abnormalities with pathogenic copy number variants in a large Chinese cohort. Ultrasound Obstet Gynecol 2022; 59:226-233. [PMID: 34090309 DOI: 10.1002/uog.23702] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/10/2020] [Revised: 05/12/2021] [Accepted: 05/21/2021] [Indexed: 06/10/2023]
Abstract
OBJECTIVES To assess the clinical utility of prenatal chromosomal microarray analysis (CMA) in fetuses with abnormal renal sonographic findings, and to evaluate the association of pathogenic or likely pathogenic copy number variants (P/LP CNVs) with different types of renal abnormality. METHODS This was a retrospective study of fetuses at 14-36 weeks screened routinely for renal and other structural abnormalities at the Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region. We retrieved and analyzed data from fetuses with abnormal renal sonographic findings, examined between January 2013 and November 2019, which underwent CMA analysis using tissue obtained from chorionic villus sampling (CVS), amniocentesis or cordocentesis. We evaluated the CMA findings according to type of renal ultrasound anomaly and according to whether renal anomalies were isolated or non-isolated. RESULTS Ten types of renal anomaly were reported on prenatal ultrasound screening, at a mean ± SD gestational age of 24.9 ± 4.8 weeks. The anomalies were diagnosed relatively late in this series, as 64% of cases with an isolated renal anomaly underwent cordocentesis rather than CVS. Fetal pyelectasis was the most common renal ultrasound finding, affecting around one-third (34.32%, 301/877) of fetuses with a renal anomaly, but only 3.65% (n = 11) of these harbored a P/LP CNV (comprising: isolated cases, 2.37% (4/169); non-isolated cases, 5.30% (7/132)). Hyperechogenic kidney was found in 5.47% (n = 48) of fetuses with a renal anomaly, of which 39.58% (n = 19) had a P/LP CNV finding (comprising: isolated cases, 44.44% (16/36); non-isolated cases, 25.00% (3/12)), the highest diagnostic yield among the different types of renal anomaly. Renal agenesis, which accounted for 9.92% (n = 87) of all abnormal renal cases, had a CMA diagnostic yield of 12.64% (n = 11) (comprising: isolated cases, 11.54% (9/78); non-isolated cases, 22.22% (2/9); unilateral cases, 11.39% (9/79); bilateral cases, 25.00% (2/8)), while multicystic dysplastic kidney (n = 110), renal cyst (n = 34), renal dysplasia (n = 27), crossed fused renal ectopia (n = 31), hydronephrosis (n = 98), renal duplication (n = 42) and ectopic kidney (n = 99) had overall diagnostic rates of 11.82%, 11.76%, 7.41%, 6.45%, 6.12%, 4.76% and 3.03%, respectively. Compared with the combined group of CMA-negative fetuses with any other type of renal anomaly, the rate of infant being alive and well at birth was significantly higher in CMA-negative fetuses with isolated fetal pyelectasis or ectopic kidney, whereas the rate was significantly lower in fetuses with isolated renal agenesis, multicystic dysplastic kidney or severe hydronephrosis. The most common pathogenic CNV was 17q12 deletion, which accounted for 30.14% (22/73) of all positive CMA findings, with a rate of 2.51% (22/877) among fetuses with an abnormal renal finding. Fetuses with 17q12 deletion exhibited a wide range of renal phenotypes. Other P/LP CNVs in the recurrent region that were associated with prenatal renal ultrasound abnormalities included 22q11.2, Xp21.1, Xp22.3, 2q13, 16p11.2 and 1q21, which, collectively, accounted for 2.17% (19/877) of the fetuses with prenatal renal anomalies. CONCLUSIONS In this retrospective review of CMA findings in a large cohort of fetuses with different types of renal ultrasound abnormality, the P/LP CNV detection rate varied significantly (3.03-39.58%) among the different types of kidney anomaly. Our data may help in the decision regarding whether to perform prenatal genetic testing in fetuses with renal ultrasound findings. Specifically, prenatal CMA testing should be performed in cases of hyperechogenic kidney, regardless of whether or not the anomaly is isolated, while it should be performed postnatally rather than prenatally in cases of fetal pyelectasis. © 2021 International Society of Ultrasound in Obstetrics and Gynecology.
Collapse
Affiliation(s)
- J Su
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - Z Qin
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - H Fu
- Department of Clinical Genetics, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Nanning, China
| | - J Luo
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - Y Huang
- Department of Ultrasound Examination, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Nanning, China
| | - P Huang
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - S Zhang
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - T Liu
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - W Lu
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - W Li
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - T Jiang
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - S Wei
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
| | - S Yang
- Department of Ultrasound Examination, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Nanning, China
| | - Y Shen
- Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Guangxi Birth Defects Prevention and Control Institute, Nanning, China
- Division of Genetics and Genomics, Boston Children's Hospital, Department of Neurology, Harvard Medical School, Boston, MA, USA
| |
Collapse
|