Automated diagnosis and prognosis of COVID-19 pneumonia from initial ER chest X-rays using deep learning

BMC Infectious Diseases

Table 1 Demographics and clinical variables of test cohort patients stratified by SARS-CoV-2 RT-PCR results

	Count	Frequency (%)	Count	Frequency (%)
Sex
Female	113	47.9	111	49.5
Male	123	52.1	109	50.5
Ethnicity
Asian	2	0.9	2	0.9
Black	112	47.5	99	45.0
Hispanic	17	7.2	0	0
Other	9	3.8	6	2.7
White	83	35.2	113	51.4
Smoking
Never	155	65.7	93	42.3
Former	19	8.1	73	33.2
Current	54	22.9	54	24.5
COPD	22	9.3	10	4.5
Cystic fibrosis	1	0.4	0	0
Asthma	32	13.6	35	15.9
Lung cancer	2	0.8	0	0
Cancer (other)	36	19.2	29	13.2
Diabetes mellitus	92	39.0	48	21.8
Hypertension	148	62.3	114	51.8
Cardiac disease	26	11.0	52	23.6
Pulmonary HTN	23	9.7	6	2.7
Sickle cell disease	6	2.5	18	8.2
Thalassemia	0	0	0	0
Organ transplant	13	5.5	4	1.8
HIV	1	0.4	4	1.8
Autoimmune	15	6.4	14	6.4
Chronic liver disease	7	3.0	8	3.6
Chronic kidney disease	48	20.3	17	7.7

SARS-CoV-2: Severe acute respiratory syndrome coronavirus 2; RT-PCR: Reverse transcription polymerase chain reaction; SD: Standard deviation; BMI: Body mass index; CXR: Chest X-ray; COPD: Chronic obstructive pulmonary disease; HTN: Hypertension; HIV: Human immunodeficiency virus

ISSN: 1471-2334