China
Africa
Explore chapters and articles related to this topic
Big data and public health
Published in Sridhar Venkatapuram, Alex Broadbent, The Routledge Handbook of Philosophy of Public Health, 2023
In addition, machine-learning algorithms are a core component of the vision of data- driven research in which theories and values play no role. Unlike conventional algorithms that merely apply rules devised by people, machine-learning algorithms can learn to detect patterns in data sets without any explicit instruction about those patterns. Some proponents of data-driven research suggest that machine-learning algorithms can therefore avoid the influence of theories and values. A common objection to this perspective highlights that our world is replete with bias and inequality, continuously shaped by the actions of people who are influenced by theories and values. Hence, even an ideal algorithm trained on a data set that perfectly reflects the world as it is would learn from a world molded by human theories and values, and so it might recapitulate existing biases and inequalities. A second objection concerns the theories and values of those who construct machine-learning algorithms, such that design choices that affect the behavior of the algorithms are constrained by the designers’ theories and values. A third objection suggests that machine-learning algorithms are constitutively theory- or value-laden: even if the prior two objections are circumvented, algorithms might require the recruiting of theories or values to overcome problems of induction and underdetermination (Johnson, forthcoming; Martin 2019).
Role of Knowledge Graphs in Analyzing Epidemics and Health Disasters
Published in Adarsh Garg, D. P. Goyal, Global Healthcare Disasters, 2023
Disease prediction is one of the major branches of the healthcare industry. Disease prediction systems aim to forecast probable disease and can even suggest precautions and cures that exist. A human doctor diagnoses several diseases based upon multiple sources of knowledge and general machine learning algorithms take into focus only the clinical data. Therefore, papers like Liu et al. suggest that multiple factors should be taken under consideration for the purpose of enhancing the precision, like medical history of the patient, rule book of the diseases, clinical health records of varied patients, experience of the experts, etc. disease prediction basically consists of three steps—collecting the data, data processing, and decision-making.
Big Data Analytics in Healthcare Data Processing
Published in Punit Gupta, Dinesh Kumar Saini, Rohit Verma, Healthcare Solutions Using Machine Learning and Informatics, 2023
Tanveer Ahmed, Rishav Singh, Ritika Singh
Veracity: Data veracity refers to the degree of certainty that a data interpretation is consistent [11]. Varied data sources have different levels of data reliability and dependability [12]. Unsupervised machine learning algorithms, on the other hand, are used in healthcare for automated machines to make decisions based on data which may be deceptive or useless [10].The purpose of healthcare analytics is to obtain useful information that can be used to make better decisions and provide better patient care.
Machine learning algorithms for integrating clinical features to predict intracranial hemorrhage in patients with acute leukemia
Published in International Journal of Neuroscience, 2023
Quanhong Chu, Wenxin Wei, Huan Lao, Yujian Li, Yafu Tan, Xiaoyong Wei, Baozi Huang, Chao Qin, Yanyan Tang
Among the enrolled acute leukemia (AL) patients, only 75 patients had ICH, there were 873 patients without ICH. As shown in Table 1, comparisons of 42 features between AL patients with ICH and without ICH were performed using Mann–Whitney U test, chi-square test or Fisher’s exact test, as appropriate. Unfortunately, those traditional statistical methods have some disadvantages such as imbalanced sample distributions, which seriously affect the statistical power [36]. However, machine learning is a fast-growing field that generates predictive or descriptive models by learning from training data rather than by being rigidly programmed [16]. Machine learning algorithms can be divided into three categories: supervised, unsupervised and semi-supervised. Supervised learning algorithms specialize in two types of problems: regression problems and classification problems, which become more practical to apply in disease risk prediction area [36]. Supervised learning algorithms include several common algorithms such as RF, LR, SVM, KNN, GNB, DT and AdaBoost. In order to identify the precise classifier to reasonably fit the 42 different features, we used different machine learning algorithm models.
Integrating artificial intelligence into an ophthalmologist’s workflow: obstacles and opportunities
Published in Expert Review of Ophthalmology, 2023
Priyal Taribagil, HD Jeffry Hogg, Konstantinos Balaskas, Pearse A Keane
There are many machine learning algorithms used in healthcare such as linear regression, logistic regression, decision trees, and random forests. None of these examples represent deep learning, which is a subset of machine learning that uses multiple layers of nodes connected in a neural network. These neural networks are able to process multiple data items, whilst preserving spatial distribution [7]. The incorporation of hidden layers aids the exploration of more complex non-linear data patterns [8]. Convolutional Neural Networks (CNN) are a subtype of neural networks used commonly in image recognition. By using multiple convolutional layers, they are able to process both simple and complex features (edges, lines, colors, shapes, etc.). Current examples of CNNs include AlexNet, GoogleNet, and ResNet[9]. Much of the successes in deep learning have been driven by CNNs.
Machine learning, pharmacogenomics, and clinical psychiatry: predicting antidepressant response in patients with major depressive disorder
Published in Expert Review of Clinical Pharmacology, 2022
William V. Bobo, Bailey Van Ommeren, Arjun P. Athreya
A common theme across all studies was the use of next-generation sequencing data from research studies, as opposed to using clinical laboratory generated gene panel data for training/testing AI/ML methods. Looking to the future, the adoption of validated machine learning approaches using genomics for predicting antidepressant response will likely require both clinical inputs (e.g. basic demographics, individual-item scores on depression rating scale, etc.) and laboratory-based inputs (e.g. SNP panel results). Therefore, novel architectures will need to be considered to run the machine learning algorithms in clinical care environments. There are at least three possible architectures suitable for this purpose: (1) a laboratory-based workflow for ordering and billing (Figure 3(a)); (2) a Clinical Decision Support electronic health record (EHR) application designed to interface directly with the ordering clinician’s EHR (Figure 3(b)); and (3) a custom platform or application that clinicians interact with directly, via a provider-facing application programming interface (API) (Figure 3(C)). Each of these architectures presents its own unique challenges to implementation, billing for service, and ongoing use by providers. Important factors to consider when selecting a model will include ordering clinicians’ requirements, regulatory concerns, timelines, information technology (IT) build complexity, the desired market, and scalability to future machine learning approaches using genomics for predicting other treatment response phenotypes for other disease groups.