|
Approaches | Year | Inventor | Method | Advantage | Disadvantage | Reference |
|
HANDX system | 1989 | Micheal and Nelson | Segmentation and isolated | Reduced observation variability | No reasonable accuracy | [58] |
PROI-based system | 1991 | Pietka et al. | Segmentation of phalanges and epiphyses | Low mean difference and error rate | Evaluated in small scale | [59] |
The CASAS system | 1994 | Tanner and Gibbons | Based on the TW2 RUS method | More accurate than manual TW method | Did not work for assessing with pathological problem | [62] |
Middle phalanx of the third finger | 2002 | Niemeijer | Segmentation of middle phalanx of third finger utilized the active shape model | Accuracy of to compared with an observer | Only covered the children between 9 and 17 years | [65] |
Neural network system based on linear | 1995 | Gross et al. | Based on linear distance measures | Better correlation coefficients | Did not use morphological feature | [66] |
Phalanges length based system | 1990 | Pietka et al. | Segmentation of phalangeal length or carpal | Reduce subjective decision | Depends on the reference population group | [67] |
The third digit-three epiphyses | 1999 | Sato et al. | Analyzing the bones of the third digit | Reasonable accuracy | Covered the children between 2 and 15 years | [68] |
Phalanges, epiphyses, and carpals | 1999 | Hsien et al. | Based on phalangeal region of interest (PROI) | Low error rate | Poor image processing techniques | [69] |
Mahmoodi model | 2000 | Mahmoodi et al. | Analysis phalangeal and active shape model | Reduced the risk in assessing the bone age by using the Bayes risk principle | Capability of further progress | [70] |
Neural network classifiers using RUS and carpal | 2008 | Liu et al. | Based on RUS and carpal bones | Small standard deviation of the differences | High image processing loading | [71] |
Neural network based on the radius and ulna | 2008 | Tristàn-Vega and Arribas | Adaptive clustering technique for segmentation | Improving the bone segmentation | Limited to four TW3 levels | [72] |
Neural network analysis based on the epiphyses and carpal | 1995 | Rucci et al. | Based on the TW method and using the epiphyses and carpal | Useful technique for classification in TW2 method. | Neural network system started in dumb state | [73] |
The royal orthopaedic hospital skeletal ageing System | 1994 | Hill and Pynsent | Based on the 13-bone and 20-bone TW2 method | Reliable method for BAA | Small group of sample images | [74] |
BoneXpert system | 2009 | Thodberg et al. | Based on shape driven and the TW RUS based | High accuracy | Rejects images in poor quality | [77] |
Web-based system using histogram | 2012 | Mansourvar et al. | Based on the histogram technique | Remove the segmentation method | Not reliable for images with poor image quality or abnormal bone structure | [78] |
|