Table 1.

Artificial intelligence models for disease diagnosis and prediction

Function	Theme	Disease type	Artificial intelligence models	Data Type	Ref.
Information extraction	Extract valuable information	Myocardial infarction	Gaussian naïve Bayes-based active balancing mechanism	Imbalanced electrocardiogram data	[8]
	Medical data processing	Beta-lactam allergy	Fast incremental decision tree	EMRs	[9]
Classification & diagnosis	Classification of chronic diseases	Chronic diseases	Hybrid deep learning	EMRs	[10]
	Classification of cardiac disorder	Cardiac disorder	Adaptive neuro-fuzzy inference system	Electrocardiogram signals	[11]
	Detect Covid-19 disease	Covid-19 disease	CNN	Chest X-ray images	[12]
	Infer illness and predict outcomes	Diabetes and mental health	LSTM	EMRs	[13]
	Brain disease prognosis	Brain disease	Weakly-supervised CNN	MRI and clinical scores	[14]
Tendency judgment	Infection rates of COVID-19	COVID-19	LSTM	EMRs	[15]
	Rehabilitation progress	Rehabilitation	CNN	Movement data	[16]
	Dynamic changes in disease	Congenital heart disease	Bayesian classification	Cardiopathy data	[17]
	Transcriptional effects of mutations	Mutations	Hybrid deep learning	DNA sequence	[18]
	Mortality detection in ICU	Unspecified	Deep learning and rule-based reasoning	EMRs	[19]
Occurrence prediction	Cognitive impairment conversion prediction	Dementia	Hybrid CNN	MRI	[20]
	Predict postoperative morbidity	Heart disease	Ensemble model	EMRs	[21]
	Predict the occurrence of a disease	Multicategory-multifactorial disease	Generalized artificial intelligence strategy	EMRs	[22]
Risk prediction	Predict the risk level of the disease	Multivariate disease	Deep learning model	EMRs	[23]
	Stratify the clinical risks of acute coronary syndrome	Acute coronary syndrome	Regularized stacked denoising auto-encoder model	EMRs	[24]
	Disease risk prediction	Unspecified	Multimodal data-based recurrent CNN	Semi-structured EMRs	[25]
	Multiple disease risk prediction	Multiple diseases	Directed disease network and recommendation system	EMRs	[26]