Deep self-supervised machine learning algorithms with a novel feature elimination and selection approaches for blood test-based multi-dimensional health risks classification.

BACKGROUND: Blood test is extensively performed for screening, diagnoses and surveillance purposes. Although it is possible to automatically evaluate the raw blood test data with the advanced deep self-supervised machine learning approaches, it has not been profoundly investigated and implemented yet. RESULTS: This paper proposes deep machine learning algorithms with multi-dimensional adaptive feature elimination, self-feature weighting and novel feature selection approaches. To classify the health risks based on the processed data with the deep layers, four machine learning algorithms having various properties from being utterly model free to gradient driven are modified. CONCLUSIONS: The results show that the proposed deep machine learning algorithms can remove the unnecessary features, assign self-importance weights, selects their most informative ones and classify the health risks automatically from the worst-case low to worst-case high values.

What Determines Levels of Mitochondrial Genetic Diversity in Birds?

What determines levels of genetic diversity in mitochondrial DNA remains unresolved. We have investigated the factors that are correlated to the level of synonymous diversity of mitochondrial DNA in more than 300 bird species. We find that diversity is significantly correlated to clutch and range size, but not significantly correlated to many other variables including body mass, latitude, and longevity. The correlation between diversity and range appears to be a consequence of a correlation between range and effective population size since a measure of the effectiveness of natural selection, which is expected to be correlated to the effective population size, is also correlated to range. The slope of the relationship between diversity and range is shallow, consistent with Lewontin's paradox, and very similar to the relationship found in mammals.

Evaluation of ALSFRS-R Scale with Fuzzy Method in Amyotrophic Lateral Sclerosis.

Introduction: Amyotrophic lateral sclerosis (ALS) is a disease with high morbidity and mortality that adversely affects the activities of daily living. Disease progression in ALS is characterized by loss of function in bulbar, motor, and respiratory parameters. The revised amyotrophic lateral sclerosis functional rating scale (ALSFRS-R), which consists of 12 criteria, is used to determine disease effects on each of these functions. While each criterion is equally important when calculating the total ALSFRS-R score, the importance levels of the 12 criteria may vary in clinical practice. In this classical approach, the relationships among the parameters are not considered and the effects of bulbar, spinal, and respiratory dysfunctions on a patient's activities of daily living may be different. Methods: In this study, we aimed to evaluate ALS cases with the ALSFRS-R fuzzy method. Although each subheading in the ALSFRS-R had the same score, the disease score was determined by the fuzzy ALSFRS-R method, based on whether a subheading had priority in management of the disease. While creating the functional rating scale ALSFRS-R approach, fuzzy ALSFRS-R score values were obtained by creating fuzzy models for each main group and integrating the fuzzy model results of each main group into a separate model. Results: In total, 50 patients with definite ALS according to the El Escorial criteria (33 men [66%] and 17 women [34%]; mean age, 58.49±10.01 years) were included in the study. When ALSFRS-R results and fuzzy ALSFRS-R results were compared, the prioritization order of 45 patients increased using the fuzzy ALSFRS-R score, while the prioritization order of five patients remained the same in both evaluations. Conclusion: The approach obtained by using fuzzy membership functions and decision rules, formed in accordance with expert opinion, was applied to the data of 50 patients from a large-scale hospital. In total, 90% of the patients had increased prioritization when using the fuzzy ALSFRS-R scoring method. Our results showed that the fuzzy approach provided more accurate information regarding a patient's condition.

Thorax CT Dose Reduction Based on Patient Features: Effect of Patient Characteristics on Image Quality and Effective Dose.

Computed tomography (CT) radiation dose reduction is vital without compromising image quality. The aim was to determine the effects of patient characteristics on the received radiation dose and image quality in chest CT examinations and to be able to predict dose and image quality prior to scanning. Consecutive 230 patients underwent routine chest CT examinations were included. CT examination and patients input parameters were recorded for each patient. The effect of patients' demographics/anthropometrics on received dose and image quality was investigated by linear regression analysis. All parameters were evaluated using an artificial neural network (ANN). Of all parameters, patient demographics/anthropometrics were found to be 98% effective in calculating dose reduction. Using ANN on 60 new patients was more than 90% accurate for output parameters and 91% for image quality. Patient characteristics have a significant impact on radiation dose and image quality. Dose and image quality can be determined before CT. This will allow setting the most appropriate scanning parameters before the CT scan.