Table 10.

Preliminary studies applying XAI to tabular data

References	# Cit.	Application	Input Data	AI model(s)	XAI method(s)	Dataset(s)
Davagdorj et al. (2021)	5	NCD prediction	EHR	DNN	SHAP	NHANES dataset${}^a$
Prentzas et al. (2019)	25	Stroke prediction	EHR	RF	InTrees, symbolic AI	ACSRS dataset Nicolaides et al. (2010)
Lisboa et al. (2020)	2	Cervical cancer diagnosis	EHR	MLP	PRN	Retrospective study
Leung et al. (2021)	11	COVID-19 diagnosis	EHR	RF, FSL-ANN	SHAP	COVID-19 dataset${}^b$
Cavaliere et al. (2020)	10	PD diagnosis	Handmade writings	Evolutionary algorithms	GE	HandPD dataset Pereira et al. (2015)
Ibrahim et al. (2020)	15	AMI prediction	ECG features	XGBoost	SHAP	ECG-ViEW II DB Sadhukhan et al. (2018)
Krishnakumar and Abdou (2020)	1	Arrhythmia detection	ECG features	stacked NB-k-NN-RF ensemble	SHAP	Arrhythmia dataset Guvenir et al. (1997)
Moreno-Sanchez (2020)	6	Survival prediction in HF patients	EHR	XGBoost	SHAP	HF dataset Ahmad et al. (2017)
Chmiel et al. (2021)	0	Hospital reattendance prediction	EHR	XGBoost	SHAP	Retrospective study
Chen et al. (2020)	15	Hospital readmission prediction in HF patients	EHR	MLP	Attention	Retrospective study
El-Bouri et al. (2020)	7	Hospital admission location prediction	EHR	ANN	Saliency maps	Retrospective study
Ochoa et al. (2021)	7	Therapy recommendation	EHR	LONN	Fuzzy rules	Pakistan database${}^c$
Costa et al. (2021)	1	Pain relief evolution assessment	EHR	RF-XGBoost ensemble	SHAP	Retrospective study
Dwivedi et al. (2021)	0	FHS detection	CTG/HRV features	LightGBM	feature permutation, SHAP	Kaggle Fetal Health dataset${}^d$
Zhai et al. (2020)	32	Sleep stage classification	Actigraphy, HR/HRV features	RF	SHAP	MESA sleep dataset Zhang et al. (2018)
Gullapalli et al. (2021)	1	Clinical opioid use tracking	HR/HRV, EDA, ST, 3-D Acc features	CTA-TCN	Attention	Pilot study

${}^a$https://www.cdc.gov/nchs/nhanes

${}^b$https://www.kaggle.com/einsteindata4u/covid19

${}^c$https://archive.ics.uci.edu/ml/datasets/Heart+failure+clinical+records

${}^d$https://www.kaggle.com/andrewmvd/fetal-health-classification