Abstract
PURPOSE: To develop a neural network to calculate skeletal age based on measurements taken from digitized hand radiographs. MATERIALS AND METHODS: From a database of 521 hand radiographs obtained in healthy patients, four parameters were calculated from seven linear measurements and were used to train a neural network, with use of the jackknife method, to calculate skeletal age. The results were compared with those of an experienced pediatric radiologist using a standard pediatric skeletal atlas. RESULTS: The mean difference from biologic age for the neural network was -0.261 years ± 1.82 (standard deviation) and for the radiologist, -0.232 years ± 1.54; this difference was not significantly different (P = .67, Wilcoxon signed rank test). Skeletal age determined by the neural network was closer to the biologic age than that assigned by the radiologist in 243 of 521 cases (47%). CONCLUSION: A simple neural network may assist radiologists in the assessment of skeletal age.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 689-695 |
| Number of pages | 7 |
| Journal | Radiology |
| Volume | 195 |
| Issue number | 3 |
| State | Published - 1995 |
Fingerprint
Keywords
- Bones, growth and development
- Children, skeletal system
- Computers, neural network
ASJC Scopus subject areas
- Radiological and Ultrasound Technology
Cite this
Pediatric skeletal age : Determination with neural networks. / Gross, G. W.; Boone, John M; Bishop, D. M.
In: Radiology, Vol. 195, No. 3, 1995, p. 689-695.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Pediatric skeletal age
T2 - Determination with neural networks
AU - Gross, G. W.
AU - Boone, John M
AU - Bishop, D. M.
PY - 1995
Y1 - 1995
N2 - PURPOSE: To develop a neural network to calculate skeletal age based on measurements taken from digitized hand radiographs. MATERIALS AND METHODS: From a database of 521 hand radiographs obtained in healthy patients, four parameters were calculated from seven linear measurements and were used to train a neural network, with use of the jackknife method, to calculate skeletal age. The results were compared with those of an experienced pediatric radiologist using a standard pediatric skeletal atlas. RESULTS: The mean difference from biologic age for the neural network was -0.261 years ± 1.82 (standard deviation) and for the radiologist, -0.232 years ± 1.54; this difference was not significantly different (P = .67, Wilcoxon signed rank test). Skeletal age determined by the neural network was closer to the biologic age than that assigned by the radiologist in 243 of 521 cases (47%). CONCLUSION: A simple neural network may assist radiologists in the assessment of skeletal age.
AB - PURPOSE: To develop a neural network to calculate skeletal age based on measurements taken from digitized hand radiographs. MATERIALS AND METHODS: From a database of 521 hand radiographs obtained in healthy patients, four parameters were calculated from seven linear measurements and were used to train a neural network, with use of the jackknife method, to calculate skeletal age. The results were compared with those of an experienced pediatric radiologist using a standard pediatric skeletal atlas. RESULTS: The mean difference from biologic age for the neural network was -0.261 years ± 1.82 (standard deviation) and for the radiologist, -0.232 years ± 1.54; this difference was not significantly different (P = .67, Wilcoxon signed rank test). Skeletal age determined by the neural network was closer to the biologic age than that assigned by the radiologist in 243 of 521 cases (47%). CONCLUSION: A simple neural network may assist radiologists in the assessment of skeletal age.
KW - Bones, growth and development
KW - Children, skeletal system
KW - Computers, neural network
UR - http://www.scopus.com/inward/record.url?scp=0029057501&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0029057501&partnerID=8YFLogxK
M3 - Article
C2 - 7753995
AN - SCOPUS:0029057501
VL - 195
SP - 689
EP - 695
JO - Radiology
JF - Radiology
SN - 0033-8419
IS - 3
ER -