scholarly journals Age Estimation from Left-Hand Radiographs with Deep Learning Methods

2021 ◽  
Vol 38 (6) ◽  
pp. 1565-1574
Author(s):  
Cüneyt Ozdemir ◽  
Mehmet Ali Gedik ◽  
Yılmaz Kaya
Keyword(s):  
Deep Learning ◽  
Forensic Medicine ◽  
Age Estimation ◽  
Criminal Liability ◽  
Bone Age ◽  
Pediatric Medicine ◽  
Require Time ◽  
Left Hand ◽  
Computer Based ◽  
Development Assessment

Bone age is estimated in pediatric medicine for medical and legal purposes. In pediatric medicine, it aids in the growth and development assessment of various diseases affecting children. In forensic medicine, it is required to determine criminal liability by age, refugee age estimation, and child-adult discrimination. In such cases, radiologists or forensic medicine specialists conduct bone age estimation from left hand-wrist radiographs using atlas methods that require time and effort. This study aims to develop a computer-based decision support system using a new modified deep learning approach to accelerate radiologists' workflow for pediatric bone age estimation from wrist radiographs. The KCRD dataset created by us was used to test the proposed method. The performance of the proposed modified IncepitonV3 model compared to IncepitonV3, MobileNetV2, EfficientNetB7 models. Acceptably high results (MAE=4.3, RMSE=5.76, and R2=0.99) were observed with the modified IncepitonV3 transfer deep learning method.

Computerized Bone Age Estimation Using Deep Learning Based Program: Evaluation of the Accuracy and Efficiency

2017 ◽  
Vol 209 (6) ◽  
pp. 1374-1380 ◽  
Author(s):  
Jeong Rye Kim ◽  
Woo Hyun Shim ◽  
Hee Mang Yoon ◽  
Sang Hyup Hong ◽  
Jin Seong Lee ◽  
...  
Keyword(s):  
Program Evaluation ◽  
Deep Learning ◽  
Age Estimation ◽  
Bone Age

Bone age estimation from carpal radiography images using deep learning

2020 ◽  
Vol 37 (6) ◽  
Author(s):  
Yih An Ding ◽  
Filipe Mutz ◽  
Klaus F. Côco ◽  
Luiz A. Pinto ◽  
Karin S. Komati
Keyword(s):  
Deep Learning ◽  
Age Estimation ◽  
Bone Age

scholarly journals Bone Age Estimation by Deep Learning in X-Ray Medical Images

2020 ◽  
Author(s):  
Behnam Kiani Kalejahi ◽  
Saeed Meshgini ◽  
Sabalan Daneshvar ◽  
Ali Farzamnia
Keyword(s):  
Deep Learning ◽  
Age Estimation ◽  
Medical Images ◽  
Bone Age ◽  
X Ray

scholarly journals Bone age estimation using deep learning and hand X-ray images

2020 ◽  
Vol 10 (3) ◽  
pp. 323-331
Author(s):  
Jang Hyung Lee ◽  
Young Jae Kim ◽  
Kwang Gi Kim
Keyword(s):  
Deep Learning ◽  
Age Estimation ◽  
Bone Age ◽  
X Ray

scholarly journals Automated age estimation of young individuals based on 3D knee MRI using deep learning

2020 ◽  
Author(s):  
Markus Auf der Mauer ◽  
Eilin Jopp-van Well ◽  
Jochen Herrmann ◽  
Michael Groth ◽  
Michael M. Morlock ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Age Estimation ◽  
Absolute Error ◽  
Machine Learning Algorithms ◽  
Chronological Age ◽  
Computer Based ◽  
Young Subjects ◽  
Magnetic Resonance Imaging Mri ◽  
Age Range

AbstractAge estimation is a crucial element of forensic medicine to assess the chronological age of living individuals without or lacking valid legal documentation. Methods used in practice are labor-intensive, subjective, and frequently comprise radiation exposure. Recently, also non-invasive methods using magnetic resonance imaging (MRI) have evaluated and confirmed a correlation between growth plate ossification in long bones and the chronological age of young subjects. However, automated and user-independent approaches are required to perform reliable assessments on large datasets. The aim of this study was to develop a fully automated and computer-based method for age estimation based on 3D knee MRIs using machine learning. The proposed solution is based on three parts: image-preprocessing, bone segmentation, and age estimation. A total of 185 coronal and 404 sagittal MR volumes from Caucasian male subjects in the age range of 13 and 21 years were available. The best result of the fivefold cross-validation was a mean absolute error of 0.67 ± 0.49 years in age regression and an accuracy of 90.9%, a sensitivity of 88.6%, and a specificity of 94.2% in classification (18-year age limit) using a combination of convolutional neural networks and tree-based machine learning algorithms. The potential of deep learning for age estimation is reflected in the results and can be further improved if it is trained on even larger and more diverse datasets.

Fully Automatic Bone Age Estimation from Left Hand MR Images

2014 ◽  
pp. 220-227 ◽  
Author(s):  
Darko Stern ◽  
Thomas Ebner ◽  
Horst Bischof ◽  
Sabine Grassegger ◽  
Thomas Ehammer ◽  
...  
Keyword(s):  
Age Estimation ◽  
Bone Age ◽  
Mr Images ◽  
Left Hand ◽  
Fully Automatic

Upper Extremity Motor Fatigability as an Early Indicator in Pediatric Onset Multiple Sclerosis

2021 ◽  
pp. 088307382199988
Author(s):  
Giuseppina Pilloni ◽  
Martin Malik ◽  
Raghav Malik ◽  
Lauren Krupp ◽  
Leigh Charvet
Keyword(s):  
Multiple Sclerosis ◽  
Grip Strength ◽  
Upper Extremity ◽  
Static Fatigue ◽  
Healthy Controls ◽  
Fatigue Index ◽  
Left Hand ◽  
Right Hand ◽  
Computer Based ◽  
Pediatric Onset

Aim: To adopt a computer-based protocol to assess grip fatigability in patients with pediatric-onset multiple sclerosis to provide detection of subtle motor involvement identifying those patients most at risk for future decline. Method: Pediatric-onset multiple sclerosis patients were recruited during routine outpatient visits to complete a grip assessment and compared to a group of healthy age- and sex-matched controls. All participants completed a computer-based measurement of standard maximal grip strength and repetitive and sustained grip performance measured by dynamic and static fatigue indices. Results: A total of 38 patients with pediatric-onset multiple sclerosis and 24 healthy controls completed the grip protocol (right-hand dominant). There were no significant group differences in maximal grip strength bilaterally (right: 21.8 vs 19.9 kg, P = .25; left: 20.4 vs 18.7 kg, P = .33), although males with pediatric-onset multiple sclerosis were significantly less strong than healthy controls (right: 26.53 vs 21.23 kg, P = .009; left; 25.13 vs 19.63 kg, P = .003). Both dynamic and static fatigue indices were significantly higher bilaterally in pediatric-onset multiple sclerosis compared with healthy control participants (left-hand dynamic fatigue index: 18.6% vs 26.7%, P = .003; right-hand static fatigue index: 28.3% vs 41.3%, P < .001; left-hand static fatigue index: 31.9% vs 42.6%, P < .001). Conclusion: Brief repeatable grip assessment including measures of dynamic and sustained static output can be a sensitive indicator of upper extremity motor involvement in pediatric-onset multiple sclerosis, potentially identifying those in need of intervention to prevent future disability.

Creation and Validation of a Shorthand Magnetic Resonance Imaging Bone Age Assessment Tool of the Knee as an Alternative Skeletal Maturity Assessment

2021 ◽  
pp. 036354652110329
Author(s):  
Cary S. Politzer ◽  
James D. Bomar ◽  
Hakan C. Pehlivan ◽  
Pradyumna Gurusamy ◽  
Eric W. Edmonds ◽  
...  
Keyword(s):  
Assessment Tool ◽  
Skeletal Maturity ◽  
Bone Age ◽  
Mri Scan ◽  
Perfect Agreement ◽  
Age Assessment ◽  
Left Hand ◽  
Knee Mri ◽  
Bone Age Assessment ◽  
Hand Radiograph

Background: In managing pediatric knee conditions, an accurate bone age assessment is often critical for diagnostic, prognostic, and treatment purposes. Historically, the Greulich and Pyle atlas (hand atlas) has been the gold standard bone age assessment tool. In 2013, a shorthand bone age assessment tool based on this atlas (hand shorthand) was devised as a simpler and more efficient alternative. Recently, a knee magnetic resonance imaging (MRI) bone age atlas (MRI atlas) was created to circumvent the need for a left-hand radiograph. Purpose: To create a shorthand version of the knee MRI atlas. Study Design: Cohort study (diagnosis); Level of evidence, 2. Methods: A shorthand bone age assessment method was created utilizing the previously published MRI atlas, which utilizes several criteria that are visualized across a series of images. The MRI shorthand draws on characteristic criteria for each age that are best observed on a single MRI scan. For validation, we performed a retrospective assessment of skeletally immature patients. One reader performed the bone age assessment using the MRI atlas and the MRI shorthand on 200 patients. Then, 4 readers performed the bone age assessment with the hand atlas, hand shorthand, MRI atlas, and MRI shorthand on a subset of 22 patients in a blinded fashion. All 22 patients had a knee MRI scan and a left-hand radiograph within 4 weeks of each other. Interobserver and intraobserver reliability, as well as variability among observers, were evaluated. Results: A total of 200 patients with a mean age of 13.5 years (range, 9.08-17.98 years) were included in this study. Also, 22 patients with a mean age of 13.3 years (range, 9.0-15.6 years) had a knee MRI scan and a left-hand radiograph within 4 weeks. The intraobserver and interobserver reliability of all 4 assessment tools were acceptable (intraclass correlation coefficient [ICC] ≥ 0.8; P < .001). When comparing the MRI shorthand with the MRI atlas, there was excellent agreement (ICC = 0.989), whereas the hand shorthand compared with the hand atlas had good agreement (ICC = 0.765). The MRI shorthand also had perfect agreement in 50% of readings among all 4 readers, and 95% of readings had agreement within 1 year, whereas the hand shorthand had perfect agreement in 32% of readings and 77% agreement within 1 year. Conclusion: The MRI shorthand is a simple and efficient means of assessing the skeletal maturity of adolescent patients with a knee MRI scan. This bone age assessment technique had interobserver and intraobserver reliability equivalent to or better than the standard method of utilizing a left-hand radiograph.

Sign in / Sign up

Export Citation Format

Share Document