The first was the Siri 2-Compartment model (W.E. Siri, 1956). It is based on the assumption that the human body is made up of just two types of material: fat-mass (FM); and fat-free-mass (FFM); and that the two are assumed to have average densities of 0.900 kg/cc and 1.100 kg/cc respectively. Further, it is assumed that the relative amounts of the three major components of the FFM (aqueous, mineral and protein) are also known as defined, and that they are constant and the same in all individuals. The Siri 2-comp model formula is Bodyfat % = (4.95/Db - 4.50) *100
Siri made an improved 3-Compartment model formula a few years later (W.E. Siri, 1961) which corrected for the total body water (Siri 3-Comp). The Siri 3-comp model formula is Bodyfat % = (2.11/Db) - (0.78 * W - 1.354) * 100
Brozek made a different 2-compartment model shortly after (J. Brozek, 1963), by revising some quantitative assumptions. The Brozek 2-comp model formula is Bodyfat % = (4.57/(Db) - 4.142) * 100
Db is the measured Hydrostatic average body density (gm/cc),
generally corrected for Residual Volume of air remaining in the lungs.
W is the total body water as a fraction of body weight.
Katch added some knowledge to the field (F.I. Katch, 1967)
In the 1980s, various researchers recognized that those formulas, especially the popular 2-compartment models, regarding gender, age, race, health, and other issues, were sometimes quite incorrect. T.G. Lohman (1984, 1986, 1989) and D.H. Nielsen (1993) specifically publishing several papers regarding application to children.
During the 1990s, several 4-compartment models were presented. Two of the most respected are the 4-Comp models of body composition referred to as the Heymsfield 4-Comp and Baumgartner 4-Comp (S.B. Heymsfield, S. Lichtman, R.N. Baumgartner, J. Wang, Y. Kamen, A. Aliprantis, and R.N. Pierson Jr., Am. Jour. Clin. Nutr. 52: 52-58, 1990.).
The Heymsfield 4-Comp model formula is Bodyfat % = {[2.748 / D(P+F) - 2.051] [BW - (A + M)]/BW} * 100
| Public Service Categories |
|
Self-Sufficiency - Many Suggestions Environmental Subjects Scientific Subjects Advanced Physics Social Subjects Religious Subjects Public Services Home Page Main Menu |
D(P+F) is the density of protein-plus-fat compartment mixtures
BW is the body weight (kg)
A is the aqueous mass (kg)
M is the mineral mass (kg)
BV is the total body volume
It is generally assumed that neutron-activation analysis provides the most complete body composition analysis currently available. However, that process is very involved and expensive, and the Heymsfield 4-Comp model is commonly now considered the criterion to evaluate other models by.
Of the simpler formulas, only the 3-compartment Siri model is generally considered of reasonable accuracy.
The primary basic research regarding actual composition seems to have all been done in the 1860s and 1870s in autopsy analysis. We suspect that the public would not tolerate newly dead human bodies being cut apart just to weigh the various parts, as being very disrespectful, which might explain why more recent data does not seem available.
A simple questionnaire could provide any such adjustments of percentages or densities for a specific individual. Gender, race, elderly, youth, big-boned, delicate, athlete, sedentary, active, muscular, etc, can then easily each initiate adjustments to appropriate parameters of the formula, for a computer or web-site program to generate accurate Bodyfat Percentage results for that individual. The questionnaire can also ask about the success of expelling air from the lungs, permitting infirm or elderly people to get reasonably accurate results without exhaling at all (where the formula would use the Average lung capacity of 2.5 liters rather than the absolute minimum of 1.2 liters in its calculations).
The basic data for the formula then is used to multiply the volume percentages by the densities of each component part, to generate a prediction for a dry body weight for the specific individual. The person weighs him/herself to establish whether the predicted dry weight is correct, which is therefore an indication that the inputted questionnaire data was accurate.
Here are the starting data figures currently used for an Average Adult person (man):
| Component | Men Volume / Volume Percentage | Women Volume / Volume Percentage | Density gm/cc |
|---|---|---|---|
| Bone | 9% (13.5% by weight) | 9% (13.5% by weight) | 1.5 |
| Blood | 7% | 7% | 1.060 |
| Muscle | 36% (38% by weight) | 30% (32% by weight) | 1.0597 |
| Air in Lungs | Residual Volume 1.2 liter / 1.4% | Residual Volume 1.2 liter / 1.4% | 0.001 |
| Gases in Intestine | Estimated 0.2 liter / 0.2% | Estimated 0.2 liter / 0.2% | 0.002 |
| Brain | 1.5 liter / 1.6% | 1.3 liter / 1.6% | 1.3-1.4 kg / 0.93-1.0 |
| Liver | 1/36 body weight / 2.8% | 1/36 body weight / 2.8% | 1.05 |
| Stomach | 1/700 body weight / 0.15% | 1/700 body weight / 0.15% | |
| Kidneys | 4.5-6 oz or 1/240 body weight or 0.4% | 4-5.5 oz or 1/240 body weight or 0.4% | |
| Spleen | 1/320-1/400 body weight / 0.3% | 1/320-1/400 body weight / 0.3% | |
| Bladder | 0.3% | 0.3% | |
| Heart | 1/169 body weight / 0.6% | 1/149 body weight / 0.7% | |
| Lungs | 1/37 body weight / 2.7% | 1/43 body weight / 2.3% | 0.345 to 0.746 |
| Skin | 4% | 4% | |
| Adipose Tissues | 5.35% | 5.65% | 0.92 |
| Intestines | 2% | 2% | |
| Pancreas | 2-3.5 oz / 0.2% | 2-3.5 oz / 0.2% | |
| Misc, cerebrospinal fluid, etc | |||
| Bodyfat | 26% (23.9% by weight) | 32% (29.4% by weight) | 0.918 |
| TOTAL | 100% | 100% | Average Body Density |
This would indicate that this woman would neutral float with a normal air load in her lungs (average density then being 0.9982 gm/cc) (in pool water that is density 0.9978 gm/cc) but would sink when she exhaled all her air. This man would sink in both cases, only being able to float with a significant in-breath of air.
The aqueous fraction of the fat-free body is slightly different for different people, being around 0.763 ±0.009 for young women, 0.723 ±0.014, for young men, 0.756 ±0.016 for elderly women, and 0.744 ±0.009 for elderly men. This suggests that slight adjustments should be made to the density figures for most of these components, due to gender and age.
index.html
index.html
C Johnson, Theoretical Physicist, Physics Degree from Univ of Chicago