METHODS AND RESULTS: We prospectively studied 3886 Asian patients (60 ± 13 years, 21% women) with HF (ejection fraction ≤40%) from 11 regions in the Asian Sudden Cardiac Death in Heart Failure study. Anaemia was defined as haemoglobin <13 g/dL (men) and <12 g/dL (women). Ethnic groups included Chinese (33.0%), Indian (26.2%), Malay (15.1%), Japanese/Korean (20.2%), and others (5.6%). Overall, anaemia was present in 41%, with a wide range across ethnicities (33-54%). Indian ethnicity, older age, diabetes, and chronic kidney disease were independently associated with higher odds of anaemia (all P
METHODS: Accident-related autopsy reports were obtained from one of the largest hospital in Kuala Lumpur. These reports belong to nine different accident-related causes of death. Master feature vector was prepared by extracting features from the collected autopsy reports by using unigram with lexical categorization. This master feature vector was used to detect cause of death [according to internal classification of disease version 10 (ICD-10) classification system] through five automated feature selection schemes, proposed expert-driven approach, five subset sizes of features, and five machine learning classifiers. Model performance was evaluated using precisionM, recallM, F-measureM, accuracy, and area under ROC curve. Four baselines were used to compare the results with the proposed system.
RESULTS: Random forest and J48 decision models parameterized using expert-driven feature selection yielded the highest evaluation measure approaching (85% to 90%) for most metrics by using a feature subset size of 30. The proposed system also showed approximately 14% to 16% improvement in the overall accuracy compared with the existing techniques and four baselines.
CONCLUSION: The proposed system is feasible and practical to use for automatic classification of ICD-10-related cause of death from autopsy reports. The proposed system assists pathologists to accurately and rapidly determine underlying cause of death based on autopsy findings. Furthermore, the proposed expert-driven feature selection approach and the findings are generally applicable to other kinds of plaintext clinical reports.
METHODS AND RESULTS: A total of 159 755 adults aged ≥35 years from Mexico City were enrolled in a prospective study and followed for 16 years. Cox regression, adjusted for confounders, yielded mortality rate ratios (RRs) associated with three markers of abdominal adiposity (waist circumference, waist-hip ratio, and waist-height ratio) and one marker of gluteo-femoral adiposity (hip circumference) for cause-specific mortality before age 75 years. To reduce reverse causality, deaths in the first 5 years of follow-up and participants with diabetes or other prior chronic disease were excluded. Among 113 163 participants without prior disease and aged 35-74 years at recruitment, all adiposity markers were positively associated with vascular-metabolic mortality. Comparing the top versus bottom tenth of the sex-specific distributions, the vascular-metabolic mortality RRs at ages 40-74 years were 2.32 [95% confidence interval (CI) 1.84-2.94] for waist circumference, 2.22 (1.71-2.88) for the waist-hip ratio, 2.63 (2.06-3.36) for the waist-height ratio, and 1.58 (1.29-1.93) for hip circumference. The RRs corresponding to each standard deviation (SD) higher usual levels of these adiposity markers were 1.34 (95% CI 1.27-1.41), 1.31 (1.23-1.39), 1.38 (1.31-1.45), and 1.18 (1.13-1.24), respectively. For the markers of abdominal adiposity, the RRs did not change much after further adjustment for other adiposity markers, but for hip circumference the association was reversed; given body mass index and waist circumference, the RR for vascular-metabolic mortality for each one SD higher usual hip circumference was 0.80 (0.75-0.86).
CONCLUSIONS: In this study of Mexican adults, abdominal adiposity (and in particular the waist-height ratio) was strongly and positively associated with vascular-metabolic mortality. For a given amount of general and abdominal adiposity, however, higher hip circumference was associated with lower vascular-metabolic mortality.
DESIGN: Data were extracted from the Malaysian Thalassaemia Registry, a web-based system accessible to enrolled users through www.mytalasemia.net.my.
SETTING: The Malaysian Thalassaemia Registry data was recorded from reports obtained from 110 participating government and university hospitals in Malaysia.
PARTICIPANTS: The patients were those attending the 110 participating hospitals for thalassaemia treatment.
INTERVENTION: Data were collected from the Malaysian Thalassaemia Registry from 2007 until the fourth quarter of 2018.
PRIMARY OUTCOME MEASURE: 7984 out of 8681 patients with thalassaemia registered in the Malaysian Thalassaemia Registry were reported alive.
RESULTS: Majority of the patients were reported in the state of Sabah (22.72%); the largest age group affected was 5.0-24.9 years old (64.45%); the largest ethnic group involved was Malay (63.95%); and the major diagnosis was haemoglobin E/β-thalassaemia (34.37%). From the 7984 patients, 56.73% were on regular blood transfusions and 61.72% were on chelation therapy. A small fraction (14.23%) has undergone splenectomy, while the percentage of patients with severe iron overload (serum ferritin ≥5000 µg/L) reduced over time. However, cardiac complications are still the main cause of death in patients with thalassaemia.
CONCLUSION: Data gathered into the registry can be used to understand the progression of the disorder, to monitor iron overload management and to improve the outcomes of treatment, to enhance preventive strategies, reduce healthcare burden and improve the quality of life. Sustainability of the Malaysian Thalassaemia Registry is important for surveillance of thalassaemia management in the country and help the national health authorities to develop more effective policies.