Expanding molecular diagnostic coverage for tuberculosis by combining computer-aided chest radiography and sputum specimen pooling: a modeling study from four high-burden countries

Table 1 Incremental and cumulative savings by country

	Tests used	Incremental savings		Cumulative savings
	N	N	%	N	%
Bangladesh
Baseline approach	24,079
CXR approach	11,448	12,631	52.5%	12,631	52.5%
Indiscriminate pooling approach	11,676	− 228	− 2.0%	12,403	51.5%
AI-guided pooling approach	9262	2414	20.7%	14,817	61.5%
Zambia
Baseline approach	2353
CXR approach	1960	393	16.7%	393	16.7%
Indiscriminate pooling approach	1352	608	31.0%	1001	42.5%
AI-guided pooling approach	1158	194	14.3%	1195	50.8%
Nigeria
Baseline approach	1021
CXR approach	738	283	27.7%	283	27.7%
Indiscriminate pooling approach	459	279	37.8%	562	55.0%
AI-guided pooling approach	434	25	5.4%	587	57.5%
Viet Nam
Baseline approach	5074
CXR approach	2917	2,157	42.5%	2157	42.5%
Indiscriminate pooling approach	2110	807	27.7%	2964	58.4%
AI-guided pooling approach	1955	155	7.3%	3119	61.5%

CXR Chest X-ray, AI Artificial intelligence. Incremental savings are the test saved compared to the previous approach. Cumulative savings are the tests saved compared to the baseline approach

ISSN: 2731-913X