Back

CT Attenuation Map Derived Body Composition Is Associated with Cardiorespiratory Fitness in Multicenter External Validation

Miller, R. J.; Yi, J.; Shanbhag, A.; Patel, K. K.; Ruddy, T. D.; Einstein, A. J.; Feher, A.; Miller, E. J.; Liang, J. X.; Ramirez, G.; Slipczuk, L.; Travin, M. I.; Alexanderson, E.; Carvajal-Juarez, I.; Packard, R. R. S.; Al-Mallah, M.; Acampa, W.; Knight, S.; Le, V. T.; Mason, S.; Wopperer, S.; Chareonthaitawee, P.; Buechel, R. R.; Rosamond, T. L.; Berman, D. S.; Dey, D.; Di Carli, M. F.; Slomka, P.

2026-05-08 cardiovascular medicine
10.64898/2026.05.07.26352573 medRxiv
Show abstract

AimsExercise capacity is a powerful predictor of cardiovascular risk. In patients unable to exercise, body composition analysis can potentially be used to estimate cardiorespiratory fitness. We developed a body composition "fitness" score, then validated its utility in two external populations. Methods and ResultsWe included patients from four sites undergoing single photon emission computed tomography (SPECT) and twelve sites undergoing positron emission tomography (PET). We quantified body composition using deep learning. We evaluated associations between body composition and good exercise capacity (defined as completing [&ge;]7 minutes on a Bruce protocol) then developed a body composition "fitness" score. We then assessed the associations of "fitness" score with exercise capacity and all-cause mortality in external populations. In total, 36471 patients were included with median age 67 (interquartile range 58 - 74). Median skeletal muscle density was higher among patients with good exercise capacity. In the external SPECT population, the body composition "fitness" score had higher prediction performance for good exercise capacity (AUC 0.771, 95% CI 0.752 - 0.789) than age (AUC 0.717, p<0.01). In the external PET population, high body composition "fitness" score was associated with lower cardiovascular death (adjusted hazard ratio 0.70, 95% CI 0.62 - 0.79, p<0.001). ConclusionsWe demonstrated that a comprehensive body composition "fitness" score could identify patients with good cardiorespiratory fitness. This approach transforms routinely acquired CT data into a surrogate marker of fitness which can be applied in patients undergoing PET, or other CT imaging, where exercise testing is not performed. Graphical AbstractOverview of study design. The overall population (n=36471) was split as outlined above. Body composition was analyzed from available computed tomography imaging. The distribution of body composition measures were analyzed in the combined single photon emission computed tomography (SPECT) populations. A body composition "fitness" score was derived to predict good exercise capacity in the internal population, with associations assessed in the two external testing populations. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=110 SRC="FIGDIR/small/26352573v1_ufig1.gif" ALT="Figure 1"> View larger version (50K): org.highwire.dtl.DTLVardef@1b98c3eorg.highwire.dtl.DTLVardef@a64282org.highwire.dtl.DTLVardef@1589559org.highwire.dtl.DTLVardef@b5423f_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

1
BMJ Open
554 papers in training set
Top 0.9%
17.7%
2
Frontiers in Cardiovascular Medicine
49 papers in training set
Top 0.3%
10.2%
3
Scientific Reports
3102 papers in training set
Top 9%
8.5%
4
Journal of Clinical Medicine
91 papers in training set
Top 0.7%
4.9%
5
PLOS ONE
4510 papers in training set
Top 38%
3.6%
6
eLife
5422 papers in training set
Top 25%
3.6%
7
npj Digital Medicine
97 papers in training set
Top 1%
3.6%
50% of probability mass above
8
Frontiers in Physiology
93 papers in training set
Top 2%
2.8%
9
Diagnostics
48 papers in training set
Top 0.6%
2.8%
10
MethodsX
14 papers in training set
Top 0.1%
2.8%
11
Journal of the American Heart Association
119 papers in training set
Top 3%
1.9%
12
Annals of Translational Medicine
17 papers in training set
Top 0.5%
1.9%
13
Journal of Translational Medicine
46 papers in training set
Top 0.9%
1.7%
14
Sensors
39 papers in training set
Top 1%
1.7%
15
The American Journal of Cardiology
15 papers in training set
Top 1.0%
1.3%
16
Computers in Biology and Medicine
120 papers in training set
Top 3%
1.2%
17
Medical Image Analysis
33 papers in training set
Top 0.8%
1.2%
18
BMC Medicine
163 papers in training set
Top 5%
1.1%
19
JAMA Network Open
127 papers in training set
Top 3%
1.0%
20
The Lancet Digital Health
25 papers in training set
Top 0.8%
1.0%
21
IEEE Transactions on Biomedical Engineering
38 papers in training set
Top 0.8%
0.9%
22
Metabolism
14 papers in training set
Top 0.4%
0.8%
23
BMJ Global Health
98 papers in training set
Top 2%
0.8%
24
Frontiers in Oncology
95 papers in training set
Top 3%
0.8%
25
Cureus
67 papers in training set
Top 5%
0.8%
26
SLAS Technology
11 papers in training set
Top 0.3%
0.8%
27
PeerJ
261 papers in training set
Top 16%
0.7%
28
Computer Methods and Programs in Biomedicine
27 papers in training set
Top 1%
0.7%
29
JMIR Research Protocols
18 papers in training set
Top 2%
0.6%
30
Ultrasound in Medicine & Biology
10 papers in training set
Top 0.7%
0.6%