Effect of knowledge of chronologic age on the variability of pediatric bone age determined using the Greulich and Pyle standards

ASSESSEMENT OF BONE AGE AGREEMENT BETWEEN THE SAUVEGRAIN AND GREULICH AND PYLE METHODS

Objective

To evaluate the intra and inter observer agreement of the Sauvegrain, Greulich and Pyle methods.

Material and methods

This is an observational, retrospective and cross-sectional study ethically approved by opinion 6,192,391. 100 radiographic images of the elbow and 100 of the left wrist and hand were collected from children whose images were selected by a researcher who did not carry out the evaluations. The Sauvegrain, Greulich and Pyle methods were used to determine bone age. We provided a detailed explanation of each method and the evaluators received a file with the study images. After three weeks, the exams were randomized and the radiograms were reevaluated. Of the 100 patients in group A, 61 (61%) were boys and 39 (39%) were girls. In group B, 67 (67%) were boys and 33 (33%) were girls.

Four statistical analyzes were used

correlation; intraclass correlation; analysis using the Bland-Altman graph; differences between groups.

Results

Intra and interobserver agreement between groups was considered excellent.

Conclusions

Despite the excellent agreement, group A presented a significantly better value than B. Biological ages show a greater difference compared to chronological ages in group A. In group B, skeletal and chronological ages do not show statistical difference according to the accuracy test. Level of Evidence III, Cross-Sectional Observational Study.

The Validation of the Greulich and Pyle Atlas for Radiological Bone Age Assessments in a Pediatric Population from the Canary Islands

Bone age assessments measure the growth and development of children and adolescents by evaluating their skeletal maturity, which is influenced by various factors like heredity, ethnicity, culture, and nutrition. The clinical standards for this assessment should be up to date and appropriate for the specific population being studied. This study validates the GP-Canary Atlas for accurately predicting bone age by analyzing posteroanterior left hand and wrist radiographs of healthy children (80 females and 134 males) from the Canary Islands across various developmental stages and genders. We found strong intra-rater reliability among all three raters, with Raters 1 and 2 indicating very high consistency (intra-class coefficients = 0.990 to 0.996) and Rater 3 displaying slightly lower but still strong reliability (intra-class coefficients = 0.921 to 0.976). The inter-rater agreement was excellent between Raters 1 and 2 but significantly lower between Rater 3 and the other two raters, with intra-class coefficients of 0.408 and 0.463 for Rater 1 and 0.327 and 0.509 for Rater 2. The accuracy analysis revealed a substantial underestimation of bone age compared to chronological age for preschool- (mean difference = 17.036 months; p < 0.001) and school-age males (mean difference = 13.298 months; p < 0.001). However, this was not observed in females, where the mean difference was minimal (3.949 months; p < 0.239). In contrast, the Atlas showed greater accuracy for teenagers, showing only a slight overestimation (mean difference = 3.159 months; p = 0.823). In conclusion, the GP-Canary Atlas demonstrates overall precision but requires caution as it underestimates the BA in preschool children and overestimates it in school-age girls and adolescents.

Enhancing skeletal age estimation accuracy using support vector regression models

OBJECTIVE

The objective of the study was to determine if support vector regression (SVR) models could enhance the accuracy of skeletal age estimation compared to original metrics.

METHOD

The study used a dataset of 5,018 individuals from Wuhan, spanning ages 1 to 17. Optimal model parameters were found using cross-validation and grid search techniques. The study compared SVR-based bone age assessment metrics with original metrics and evaluated the performance of the SVR model across different sample sizes.

RESULTS

The findings unequivocally demonstrated SVR's superior reliability over original metrics in assessing bone age among children in central China. Regardless of the training set size, constructing SVR models based on TW3, CHN05, or a combination of TW3, CHN05, and GP consistently results in top-tier predictive accuracy.

CONCLUSION

This research highlights SVR's potential for accuracy improvement and robustness with limited datasets.

Association between clinical history in the radiographic request and diagnostic accuracy of thorax radiographs in dogs: A retrospective case-control study

BACKGROUND

The effect of clinical history on the interpretation of radiographs has been widely researched in human medicine. There is, however, no data on this topic in veterinary medicine.

HYPOTHESIS/OBJECTIVES

Diagnostic accuracy would improve when history was supplied.

ANIMALS

Thirty client-owned dogs with abnormal findings on thoracic radiographs and confirmation of the disease, and 30 healthy client-owned controls were drawn retrospectively.

METHODS

Retrospective case-control study. Sixty radiographic studies of the thorax were randomized and interpreted by 6 radiologists; first, with no access to the clinical information; and a second time with access to all pertinent clinical information and signalment.

RESULTS

A significant increase in diagnostic accuracy was noted when clinical information was provided (64.4% without and 75.2% with clinical information; P = .002). There was no significant difference in agreement between radiologists when comparing no clinical information and with clinical information (Kappa 0.313 and 0.300, respectively).

CONCLUSIONS AND CLINICAL IMPORTANCE

The addition of pertinent clinical information to the radiographic request significantly improves the diagnostic accuracy of thorax radiographs of dogs and is recommended as standard practice.

Validation of a deep learning algorithm for bone age estimation among patients in the city of São Paulo, Brazil

Objective

To validate a deep learning (DL) model for bone age estimation in individuals in the city of São Paulo, comparing it with the Greulich and Pyle method.

Materials and Methods

This was a cross-sectional study of hand and wrist radiographs obtained for the determination of bone age. The manual analysis was performed by an experienced radiologist. The model used was based on a convolutional neural network that placed third in the 2017 Radiological Society of North America challenge. The mean absolute error (MAE) and the root-mean-square error (RMSE) were calculated for the model versus the radiologist, with comparisons by sex, race, and age.

Results

The sample comprised 714 examinations. There was a correlation between the two methods, with a coefficient of determination of 0.94. The MAE of the predictions was 7.68 months, and the RMSE was 10.27 months. There were no statistically significant differences between sexes or among races (p > 0.05). The algorithm overestimated bone age in younger individuals (p = 0.001).

Conclusion

Our DL algorithm demonstrated potential for estimating bone age in individuals in the city of São Paulo, regardless of sex and race. However, improvements are needed, particularly in relation to its use in younger patients.

Forensic age estimation in living children: how accurate is the Greulich-Pyle method in Sabah, East Malaysia?

Background

The Greulich and Pyle's Radiographic Atlas of Skeletal Development of the Hand and Wrist (GP Atlas) is the most widely used method of determining the bone age (BA) of a child. It is also a widely accepted method for forensic age determination. As there is limited local bone age data for forensic age estimation, the purpose of this study was to assess the accuracy of the GP Atlas for forensic age determination in living Sabahan children.

Method

This study recruited 182 children between the ages of 9 years to 18 years. BA estimation of the left-hand anteroposterior radiographs were performed by two experienced radiologists using the Greulich-Pyle method.

Results

The BA estimates from two radiologists had very high interobserver reliability (ICC 0.937) and a strong positive interobserver correlation (r > 0.90). The GP method, significantly and consistently underestimated chronological age (CA) by 0.7, 0.6 and 0.7 years in overall children, boys and girls respectively with minimal errors. Mean absolute error and root of mean squared error for overall children was 1.5 and 2.2 years respectively, while mean absolute percentage error was 11.6%. This underestimation was consistent across all age groups but was statistically significant only at 13-13.9 and 17-18.9 years old age groups.

Conclusion

Despite high interobserver reliability of BA estimation using the GP Atlas, this method consistently underestimates the age of the child in all children to a significant degree, for both boys and girls across all age groups, with an acceptably low level of error metrics. Our findings suggest that locally validated GP Atlas or other type of assessments (artificial intelligence or machine learning) are needed for assessment of BA to accurately predict CA, since current GP Atlas standards significantly underestimated chronological age with minimal error for children in Sabah. A larger population-based study would be necessary for establishing a validated atlas of a bone age in Malaysia.

Bone age recognition based on mask R-CNN using xception regression model

Background and Objective: Bone age detection plays an important role in medical care, sports, judicial expertise and other fields. Traditional bone age identification and detection is according to manual interpretation of X-ray images of hand bone by doctors. This method is subjective and requires experience, and has certain errors. Computer-aided detection can effectually enhance the validity of medical diagnosis, especially with the fast development of machine learning and neural network, the method of bone age recognition using machine learning has gradually become the focus of research, which has the advantages of simple data pretreatment, good robustness and high recognition accuracy. Methods: In this paper, the hand bone segmentation network based on Mask R-CNN was proposed to segment the hand bone area, and the segmented hand bone region was directly input into the regression network for bone age evaluation. The regression network is using an enhancd network Xception of InceptionV3. After the output of Xception, the convolutional block attention module is connected to refine the feature mapping from channel and space to obtain more effective features. Results: According to the experimental results, the hand bone segmentation network model based on Mask R-CNN can segment the hand bone region and eliminate the interference of redundant background information. The average Dice coefficient on the verification set is 0.976. The mean absolute error of predicting bone age on our data set was only 4.97 months, which exceeded the accuracy of most other bone age assessment methods. Conclusion: Experiments show that the accuracy of bone age assessment can be enhancd by using the Mask R-CNN-based hand bone segmentation network and the Xception bone age regression network to form a model, which can be well applied to actual clinical bone age assessment.

Generalizability and Bias in a Deep Learning Pediatric Bone Age Prediction Model Using Hand Radiographs

Background Although deep learning (DL) models have demonstrated expert-level ability for pediatric bone age prediction, they have shown poor generalizability and bias in other use cases. Purpose To quantify generalizability and bias in a bone age DL model measured by performance on external versus internal test sets and performance differences between different demographic groups, respectively. Materials and Methods The winning DL model of the 2017 RSNA Pediatric Bone Age Challenge was retrospectively evaluated and trained on 12 611 pediatric hand radiographs from two U.S. hospitals. The DL model was tested from September 2021 to December 2021 on an internal validation set and an external test set of pediatric hand radiographs with diverse demographic representation. Images reporting ground-truth bone age were included for study. Mean absolute difference (MAD) between ground-truth bone age and the model prediction bone age was calculated for each set. Generalizability was evaluated by comparing MAD between internal and external evaluation sets with use of t tests. Bias was evaluated by comparing MAD and clinically significant error rate (rate of errors changing the clinical diagnosis) between demographic groups with use of t tests or analysis of variance and χ² tests, respectively (statistically significant difference defined as P < .05). Results The internal validation set had images from 1425 individuals (773 boys), and the external test set had images from 1202 individuals (mean age, 133 months ± 60 [SD]; 614 boys). The bone age model generalized well to the external test set, with no difference in MAD (6.8 months in the validation set vs 6.9 months in the external set; P = .64). Model predictions would have led to clinically significant errors in 194 of 1202 images (16%) in the external test set. The MAD was greater for girls than boys in the internal validation set (P = .01) and in the subcategories of age and Tanner stage in the external test set (P < .001 for both). Conclusion A deep learning (DL) bone age model generalized well to an external test set, although clinically significant sex-, age-, and sexual maturity-based biases in DL bone age were identified. © RSNA, 2022 Online supplemental material is available for this article See also the editorial by Larson in this issue.

Validation of automated bone age analysis from hand radiographs in a North American pediatric population

BACKGROUND

Radiographic bone age assessment by automated software is precise and instantaneous.

OBJECTIVE

The aim of this study was to evaluate the accuracy of an automated tool for bone age assessment.

MATERIALS AND METHODS

We compared a total of 586 bone age radiographs from 451 patients, which had been assessed by three radiologists from 2013 to 2018, with bone age analysis by BoneXpert, using the Greulich and Pyle method. We made bone age comparisons in different patient groups based on gender, diagnosis and race, and in a subset with repeated bone age studies. We calculated Spearman correlation (r) and accuracy (root mean square error, or R²).

RESULTS

Bone age analyses by automated and manual assessments showed a strong correlation (r=0.98; R²=0.96; P<0.0001), with the mean bone age difference of 0.12±0.76 years. Bone age comparisons by the two methods remained strongly correlated (P<0.0001) when stratified by gender, common endocrine conditions including growth disorders and early/precocious puberty, and race. In the longitudinal analysis, we also found a strong correlation between the automated software and manual bone age over time (r=0.7852; R²=0.63; P<0.01).

CONCLUSION

Automated bone age assessment was found to be reliable and accurate in a large cohort of pediatric patients in a clinical practice setting in North America.

Clinical application of artificial intelligence in longitudinal image analysis of bone age among GHD patients

OBJECTIVE

This study aims to explore the clinical value of artificial intelligence (AI)-assisted bone age assessment (BAA) among children with growth hormone deficiency (GHD).

METHODS

A total of 290 bone age (BA) radiographs were collected from 52 children who participated in the study at Sun Yat-sen Memorial Hospital between January 2016 and August 2017. Senior pediatric endocrinologists independently evaluated BA according to the China 05 (CH05) method, and their consistent results were regarded as the gold standard (GS). Meanwhile, two junior pediatric endocrinologists were asked to assessed BA both with and without assistance from the AI-based BA evaluation system. Six months later, around 20% of the images assessed by the junior pediatric endocrinologists were randomly selected to be re-evaluated with the same procedure half a year later. Root mean square error (RMSE), mean absolute error (MAE), accuracy, and Bland-Altman plots were used to compare differences in BA. The intra-class correlation coefficient (ICC) and one-way repeated ANOVA were used to assess inter- and intra-observer variabilities in BAA. A boxplot of BA evaluated by different raters during the course of treatment and a mixed linear model were used to illustrate inter-rater effect over time.

RESULTS

A total of 52 children with GHD were included, with mean chronological age and BA by GS of 6.64 ± 2.49 and 5.85 ± 2.30 years at baseline, respectively. After incorporating AI assistance, the performance of the junior pediatric endocrinologists improved (P < 0.001), with MAE and RMSE both decreased by more than 1.65 years (Rater 1: ΔMAE = 1.780, ΔRMSE = 1.655; Rater 2: ΔMAE = 1.794, ΔRMSE = 1.719), and accuracy increasing from approximately 10% to over 91%. The ICC also increased from 0.951 to 0.990. During GHD treatment (at baseline, 6-, 12-, 18-, and 24-months), the difference decreased sharply when AI was applied. Furthermore, a significant inter-rater effect (P = 0.002) also vanished upon AI involvement.

CONCLUSION

AI-assisted interpretation of BA can improve accuracy and decrease variability in results among junior pediatric endocrinologists in longitudinal cohort studies, which shows potential for further clinical application.

Autonomous artificial intelligence in pediatric radiology: the use and perception of BoneXpert for bone age assessment

BACKGROUND

The autonomous artificial intelligence (AI) system for bone age rating (BoneXpert) was designed to be used in clinical radiology practice as an AI-replace tool, replacing the radiologist completely.

OBJECTIVE

The aim of this study was to investigate how the tool is used in clinical practice. Are radiologists more inclined to use BoneXpert to assist rather than replace themselves, and how much time is saved?

MATERIALS AND METHODS

We sent a survey consisting of eight multiple-choice questions to 282 radiologists in departments in Europe already using the software.

RESULTS

The 97 (34%) respondents came from 18 countries. Their answers revealed that before installing the automated method, 83 (86%) of the respondents took more than 2 min per bone age rating; this fell to 20 (21%) respondents after installation. Only 17/97 (18%) respondents used BoneXpert to completely replace the radiologist; the rest used it to assist radiologists to varying degrees. For instance, 39/97 (40%) never overruled the automated reading, while 9/97 (9%) overruled more than 5% of the automated ratings. The majority 58/97 (60%) of respondents checked the radiographs themselves to exclude features of underlying disease.

CONCLUSION

BoneXpert significantly reduces reporting times for bone age determination. However, radiographic analysis involves more than just determining bone age. It also involves identification of abnormalities, and for this reason, radiologists cannot be completely replaced. AI systems originally developed to replace the radiologist might be more suitable as AI assist tools, particularly if they have not been validated to work autonomously, including the ability to omit ratings when the image is outside the range of validity.

Clinical Validation of a Deep Learning-Based Hybrid (Greulich-Pyle and Modified Tanner-Whitehouse) Method for Bone Age Assessment

OBJECTIVE

To evaluate the accuracy and clinical efficacy of a hybrid Greulich-Pyle (GP) and modified Tanner-Whitehouse (TW) artificial intelligence (AI) model for bone age assessment.

MATERIALS AND METHODS

A deep learning-based model was trained on an open dataset of multiple ethnicities. A total of 102 hand radiographs (51 male and 51 female; mean age ± standard deviation = 10.95 ± 2.37 years) from a single institution were selected for external validation. Three human experts performed bone age assessments based on the GP atlas to develop a reference standard. Two study radiologists performed bone age assessments with and without AI model assistance in two separate sessions, for which the reading time was recorded. The performance of the AI software was assessed by comparing the mean absolute difference between the AI-calculated bone age and the reference standard. The reading time was compared between reading with and without AI using a paired t test. Furthermore, the reliability between the two study radiologists' bone age assessments was assessed using intraclass correlation coefficients (ICCs), and the results were compared between reading with and without AI.

RESULTS

The bone ages assessed by the experts and the AI model were not significantly different (11.39 ± 2.74 years and 11.35 ± 2.76 years, respectively, p = 0.31). The mean absolute difference was 0.39 years (95% confidence interval, 0.33-0.45 years) between the automated AI assessment and the reference standard. The mean reading time of the two study radiologists was reduced from 54.29 to 35.37 seconds with AI model assistance (p < 0.001). The ICC of the two study radiologists slightly increased with AI model assistance (from 0.945 to 0.990).

CONCLUSION

The proposed AI model was accurate for assessing bone age. Furthermore, this model appeared to enhance the clinical efficacy by reducing the reading time and improving the inter-observer reliability.

A Novel Shorthand Approach to Knee Bone Age Using MRI: A Validation and Reliability Study

Background:

Bone-age determination remains a difficult process. An atlas for bone age has been created from knee-ossification patterns on magnetic resonance imaging (MRI), thereby avoiding the need for radiographs and associated costs, radiation exposure, and clinical inefficiency. Shorthand methods for bone age can be less time-consuming and require less extensive training as compared with conventional methods.

Purpose:

To create and validate a novel shorthand algorithm for bone age based on knee MRIs that could correlate with conventional hand bone age and demonstrate reliability across medical trainees.

Study Design:

Cohort study (diagnosis); Level of evidence, 2.

Methods:

Included in this study were adolescent patients who underwent both knee MRI and hand bone age radiographs within 90 days between 2009 and 2018. A stepwise algorithm for predicting bone age using knee MRI was developed separately for male and female patients, and 7 raters at varying levels of training used the algorithm to determine the bone age for each MRI. The shorthand algorithm was validated using Spearman rho (r_S) to correlate each rater’s predicted MRI bone age with the recorded Greulich and Pyle (G&P) hand bone age. Interrater and intrarater reliability were also calculated using intraclass correlation coefficients (ICCs).

Results:

A total of 38 patients (44.7% female) underwent imaging at a mean age of 12.8 years (range, 9.3-15.7 years). Shorthand knee MRI bone age scores were strongly correlated with G&P hand bone age (r_S = 0.83; P < .001). The shorthand algorithm was a valid predictor of G&P hand bone age regardless of level of training, as medical students (r_S = 0.75), residents (r_S = 0.81), and attending physicians (r_S = 0.84) performed similarly. The interrater reliability of our shorthand algorithm was 0.81 (95% CI, 0.73-0.88), indicating good to excellent interobserver agreement. Respondents also demonstrated consistency, with 6 of 7 raters demonstrating excellent intrarater reliability (median ICC, 0.86 [range, 0.68-0.96]).

Conclusion:

This shorthand algorithm is a consistent, reliable, and valid way to determine skeletal maturity using knee MRI in patients aged 9 to 16 years and can be utilized across different levels of orthopaedic and radiographic expertise. This method is readily applicable in a clinical setting and may reduce the need for routine hand bone age radiographs.

Skeletal bone age prediction based on a deep residual network with spatial transformer

OBJECTIVE

Bone age prediction can be performed by medical experts manually assessment of X-ray images of the hand bone. In practice, the workload is huge, resource consumption is large, measurement takes a long time, and it is easily influenced by human factors. As such, manual estimation of bone age takes a long time and the results fluctuate greatly depending on the proficiency of the radiologist.

METHODS

The left-hand X-ray image data was identified and pre-processed. X-ray image analysis method using on deep neural network was used to automatically extract the key features of the left-hand joint bone age, and evaluation performance of the model was implemented.

RESULTS

In this paper, the deep learning method can be used to obtain the X-ray bone image features, and the convolutional neural network is used to automatically assess the age of bone. The feature region extraction method based on deep learning can extract feature information with superior performance compared to the traditional image analysis technique. Based on the residual network (ResNet) model in the deep learning algorithm, the average absolute error of the age of bones detected by the bone age assessment model is 0.455 better than traditional methods and only end-to-end deep learning methods. When the learning rate is greater than 0.0005, the MAE of Inception Resnet v2 model is higher than most models. Accuracy of bone age prediction is as high as 97.6%.

CONCLUSION

In comparison with the traditional machine learning feature extraction technique, the convolutional neural network based on feature extraction has better performance in the bone age regression model, and further improves the accuracy of image-based age of bone assessment.

Bone Age Assessment Empowered with Deep Learning: A Survey, Open Research Challenges and Future Directions

Deep learning is a quite useful and proliferating technique of machine learning. Various applications, such as medical images analysis, medical images processing, text understanding, and speech recognition, have been using deep learning, and it has been providing rather promising results. Both supervised and unsupervised approaches are being used to extract and learn features as well as for the multi-level representation of pattern recognition and classification. Hence, the way of prediction, recognition, and diagnosis in various domains of healthcare including the abdomen, lung cancer, brain tumor, skeletal bone age assessment, and so on, have been transformed and improved significantly by deep learning. By considering a wide range of deep-learning applications, the main aim of this paper is to present a detailed survey on emerging research of deep-learning models for bone age assessment (e.g., segmentation, prediction, and classification). An enormous number of scientific research publications related to bone age assessment using deep learning are explored, studied, and presented in this survey. Furthermore, the emerging trends of this research domain have been analyzed and discussed. Finally, a critical discussion section on the limitations of deep-learning models has been presented. Open research challenges and future directions in this promising area have been included as well.

Rethinking Greulich and Pyle: A Deep Learning Approach to Pediatric Bone Age Assessment Using Pediatric Trauma Hand Radiographs

Purpose

To develop a deep learning approach to bone age assessment based on a training set of developmentally normal pediatric hand radiographs and to compare this approach with automated and manual bone age assessment methods based on Greulich and Pyle (GP).

Methods

In this retrospective study, a convolutional neural network (trauma hand radiograph-trained deep learning bone age assessment method [TDL-BAAM]) was trained on 15 129 frontal view pediatric trauma hand radiographs obtained between December 14, 2009, and May 31, 2017, from Children's Hospital of New York, to predict chronological age. A total of 214 trauma hand radiographs from Hasbro Children's Hospital were used as an independent test set. The test set was rated by the TDL-BAAM model as well as a GP-based deep learning model (GPDL-BAAM) and two pediatric radiologists (radiologists 1 and 2) using the GP method. All ratings were compared with chronological age using mean absolute error (MAE), and standard concordance analyses were performed.

Results

The MAE of the TDL-BAAM model was 11.1 months, compared with 12.9 months for GPDL-BAAM (P = .0005), 14.6 months for radiologist 1 (P < .0001), and 16.0 for radiologist 2 (P < .0001). For TDL-BAAM, 95.3% of predictions were within 24 months of chronological age compared with 91.6% for GPDL-BAAM (P = .096), 86.0% for radiologist 1 (P < .0001), and 84.6% for radiologist 2 (P < .0001). Concordance was high between all methods and chronological age (intraclass coefficient > 0.93). Deep learning models demonstrated a systematic bias with a tendency to overpredict age for younger children versus radiologists who showed a consistent mean bias.

Conclusion

A deep learning model trained on pediatric trauma hand radiographs is on par with automated and manual GP-based methods for bone age assessment and provides a foundation for developing population-specific deep learning algorithms for bone age assessment in modern pediatric populations.Supplemental material is available for this article.© RSNA, 2020See also the commentary by Halabi in this issue.

Bone age determination using only the index finger: a novel approach using a convolutional neural network compared with human radiologists

BACKGROUND

Recently developed convolutional neural network (CNN) models determine bone age more accurately than radiologists.

OBJECTIVE

The purpose of this study was to determine whether a CNN and radiologists can accurately predict bone age from radiographs using only the index finger rather than the whole hand.

MATERIALS AND METHODS

We used a public anonymized dataset provided by the Radiological Society of North America (RSNA) pediatric bone age challenge. The dataset contains 12,611 hand radiographs for training and 200 radiographs for testing. The index finger was cropped from these images to create a second dataset. Separate CNN models were trained using the whole-hand radiographs and the cropped second-digit dataset using the consensus ground truth provided by the RSNA bone age challenge. Bone age determination using both models was compared with ground truth as provided by the RSNA dataset. Separately, three pediatric radiologists determined bone age from the whole-hand and index-finger radiographs, and the consensus was compared to the ground truth and CNN-model-determined bone ages.

RESULTS

The mean absolute difference between the ground truth and CNN bone age for whole-hand and index-finger was similar (4.7 months vs. 5.1 months, P=0.14), and both values were significantly smaller than that for radiologist bone age determination from the single-finger radiographs (8.0 months, P<0.0001).

CONCLUSION

CNN-model-determined bone ages from index-finger radiographs are similar to whole-hand bone age interpreted by radiologists in the dataset, as well as a model trained on the whole-hand radiograph. In addition, the index-finger model performed better than the ground truth compared to subspecialty trained pediatric radiologists also using only the index finger to determine bone age. The radiologist interpreting bone age can use the second digit as a reliable starting point in their search pattern.

Incorporated region detection and classification using deep convolutional networks for bone age assessment

Bone age assessment plays an important role in the endocrinology and genetic investigation of patients. In this paper, we proposed a deep learning-based approach for bone age assessment by integration of the Tanner-Whitehouse (TW3) methods and deep convolution networks based on extracted regions of interest (ROI)-detection and classification using Faster-RCNN and Inception-v4 networks, respectively. The proposed method allows exploration of expert knowledge from TW3 and features engineering from deep convolution networks to enhance the accuracy of bone age assessment. The experimental results showed a mean absolute error of about 0.59 years between expert radiologists and the proposed method, which is the best performance among state-of-the-art methods.

MABAL: a Novel Deep-Learning Architecture for Machine-Assisted Bone Age Labeling

Bone age assessment (BAA) is a commonly performed diagnostic study in pediatric radiology to assess skeletal maturity. The most commonly utilized method for assessment of BAA is the Greulich and Pyle method (Pediatr Radiol 46.9:1269-1274, 2016; Arch Dis Child 81.2:172-173, 1999) atlas. The evaluation of BAA can be a tedious and time-consuming process for the radiologist. As such, several computer-assisted detection/diagnosis (CAD) methods have been proposed for automation of BAA. Classical CAD tools have traditionally relied on hard-coded algorithmic features for BAA which suffer from a variety of drawbacks. Recently, the advent and proliferation of convolutional neural networks (CNNs) has shown promise in a variety of medical imaging applications. There have been at least two published applications of using deep learning for evaluation of bone age (Med Image Anal 36:41-51, 2017; JDI 1-5, 2017). However, current implementations are limited by a combination of both architecture design and relatively small datasets. The purpose of this study is to demonstrate the benefits of a customized neural network algorithm carefully calibrated to the evaluation of bone age utilizing a relatively large institutional dataset. In doing so, this study will aim to show that advanced architectures can be successfully trained from scratch in the medical imaging domain and can generate results that outperform any existing proposed algorithm. The training data consisted of 10,289 images of different skeletal age examinations, 8909 from the hospital Picture Archiving and Communication System at our institution and 1383 from the public Digital Hand Atlas Database. The data was separated into four cohorts, one each for male and female children above the age of 8, and one each for male and female children below the age of 10. The testing set consisted of 20 radiographs of each 1-year-age cohort from 0 to 1 years to 14-15+ years, half male and half female. The testing set included left-hand radiographs done for bone age assessment, trauma evaluation without significant findings, and skeletal surveys. A 14 hidden layer-customized neural network was designed for this study. The network included several state of the art techniques including residual-style connections, inception layers, and spatial transformer layers. Data augmentation was applied to the network inputs to prevent overfitting. A linear regression output was utilized. Mean square error was used as the network loss function and mean absolute error (MAE) was utilized as the primary performance metric. MAE accuracies on the validation and test sets for young females were 0.654 and 0.561 respectively. For older females, validation and test accuracies were 0.662 and 0.497 respectively. For young males, validation and test accuracies were 0.649 and 0.585 respectively. Finally, for older males, validation and test set accuracies were 0.581 and 0.501 respectively. The female cohorts were trained for 900 epochs each and the male cohorts were trained for 600 epochs. An eightfold cross-validation set was employed for hyperparameter tuning. Test error was obtained after training on a full data set with the selected hyperparameters. Using our proposed customized neural network architecture on our large available data, we achieved an aggregate validation and test set mean absolute errors of 0.637 and 0.536 respectively. To date, this is the best published performance on utilizing deep learning for bone age assessment. Our results support our initial hypothesis that customized, purpose-built neural networks provide improved performance over networks derived from pre-trained imaging data sets. We build on that initial work by showing that the addition of state-of-the-art techniques such as residual connections and inception architecture further improves prediction accuracy. This is important because the current assumption for use of residual and/or inception architectures is that a large pre-trained network is required for successful implementation given the relatively small datasets in medical imaging. Instead we show that a small, customized architecture incorporating advanced CNN strategies can indeed be trained from scratch, yielding significant improvements in algorithm accuracy. It should be noted that for all four cohorts, testing error outperformed validation error. One reason for this is that our ground truth for our test set was obtained by averaging two pediatric radiologist reads compared to our training data for which only a single read was used. This suggests that despite relatively noisy training data, the algorithm could successfully model the variation between observers and generate estimates that are close to the expected ground truth.

Automatic Determination of the Greulich-Pyle Bone Age as an Alternative Approach for Chinese Children with Discordant Bone Age

BACKGROUND

Automated bone age (BA) rating using BoneXpert is being adopted worldwide. This study investigated whether manual matching of hand radiographs could be replaced by BoneXpert for BA ratings of Chinese children with delayed or advanced BA.

METHODS

482 left-hand radiographs from 482 children (aged 2-16 years) with discordant BA were evaluated by BoneXpert and manually by 4 radiology residents using the Greulich and Pyle atlas. Radiographs whose BoneXpert BA deviated by >1 year from manual assessment were rerated by 2 attending radiologists in a blinded manner.

RESULTS

Among all 482 radiographs, 46 (9.5%) radiographs were rerated and no radiographs were rejected. Differences between BoneXpert and manual rating of 28 (5.8%) cases were >1 year. The manual BAs of the 28 radiographs were all >10 years and greater than the BoneXpert BAs. The root mean square deviation between the residents and BoneXpert was 0.56 for these children (95% CI 0.53-0.61).

CONCLUSION

BoneXpert agreed with manual BA rating in 94.2% of the images. Therefore, BoneXpert could be used as an alternative for the radiology residents to make an initial BA estimation. Modification of BoneXpert should provide greater accuracy for the estimation of BA in children aged >10 years with discordant BA.

Skeletal development of the hand and wrist: digital bone age companion-a suitable alternative to the Greulich and Pyle atlas for bone age assessment?

PURPOSE

To assess reader performance and subjective workflow experience when reporting bone age studies with a digital bone age reference as compared to the Greulich and Pyle atlas (G&P). We hypothesized that pediatric radiologists would achieve equivalent results with each method while digital workflow would improve speed, experience, and reporting quality.

MATERIALS AND METHODS

IRB approval was obtained for this HIPAA-compliant study. Two pediatric radiologists performed research interpretations of bone age studies randomized to either the digital (Digital Bone Age Companion, Oxford University Press) or G&P method, generating reports to mimic clinical workflow. Bone age standard selection, interpretation-reporting time, and user preferences were recorded. Reports were reviewed for typographical or speech recognition errors. Comparisons of agreement were conducted by way of Fisher's exact tests. Interpretation-reporting times were analyzed on the natural logarithmic scale via a linear mixed model and transformed to the geometric mean. Subjective workflow experience was compared with an exact binomial test. Report errors were compared via a paired random permutation test.

RESULTS

There was no difference in bone age determination between atlases (p = 0.495). The interpretation-reporting time (p < 0.001) was significantly faster with the digital method. The faculty indicated preference for the digital atlas (p < 0.001). Signed reports had fewer errors with the digital atlas (p < 0.001).

CONCLUSIONS

Bone age study interpretations performed with the digital method were similar to those performed with the Greulich and Pyle atlas. The digital atlas saved time, improved workflow experience, and reduced reporting errors relative to the Greulich and Pyle atlas when integrated into electronic workflow.

Use of Automated Bone Age for Critical Growth Assessment

BACKGROUND

We compared the variability of bone age (BA) rating between clinicians and an automated system in children with congenital adrenal hyperplasia (CAH).

METHOD

A total of 126 radiographs assessed by 3 clinicians blinded to patient details using Greulich and Pyle (GP) (readers 1, 2, and 3) and BoneXpert (BX).

RESULTS

Comparing BA rating with each other, the mean of the absolute differences varied from 0.42 ± 0.53 years (reader 1 and BX) to 0.57 ± 0.58 years (reader 2 and reader 3), P = .368. Comparing ratings that were consistent with all 4 methods (within 1 year of each other, 93/126, 74%) and the remaining, "outliers" (33/126, 26%), the outliers were younger (P = .003), smaller (height, P = .011, weight, P = .000), and prepubertal (P = .001).

CONCLUSION

The variability of BA rating in CAH children is similar whether performed by clinicians or an automated system. The greatest variability was in prepubertal children.

Is Greulich and Pyle standards of skeletal maturation applicable for age estimation in South Indian Andhra children?

Introduction:

Now-a-day age determination has gained importance for various forensic and legal reasons. Skeletal age (SA) of a test population can be estimated by comparing with established standards of Greulich and Pyle (G-P). As this atlas has been prepared using data from upper-class children born between 1917 and 1942 in the USA and the applicability of these standards to contemporary populations has yet to be tested on Andhra children living in India. Hence, this study was aimed to assess the reliability of bone age calculated by G-P atlas in estimation of age in selected population.

Materials and Methods:

A total of 660 children (330 girls, 330 boys) between ages 9 and 20 years were randomly selected from outpatient Department of Oral Medicine in GITAM Dental College, Andhra Pradesh. Digital hand-wrist radiographs were obtained and assessed for SA using G-P atlas and the difference between estimated SA and chronological age (CA) were compared with paired t-test and Wilcoxon signed rank test.

Results:

G-P method underestimated the SA by 0.23 ± 1.53 years for boys and overestimated SA by 0.02 ± 2 years in girls and mild underestimation was noted in the total sample of about 0.1 ± 1.78 years. Spearman rank test showed significant correlation between SA and CA (r = 0.86; P < 0.001).

Conclusion:

This study concluded that G-P standards were reliable in assessing age in South Indian Andhra children of age 9–20 years with unknown CA.

Automated Greulich-Pyle bone age determination in children with chronic kidney disease

BACKGROUND

Growth restriction and retarded bone age are common findings in children with chronic kidney disease (CKD). We compared the automated BoneXpert™ method with the manual assessment of an X-ray of the non-dominant hand.

METHODS

In this retrospective multicenter study, 359 patients with CKD stages 2-5, aged 2-14.5 (girls) or 2.5-17 years (boys) were included. Bone age was determined manually by three experts (according to Greulich and Pyle). Automated determination of bone age was performed using the image analysis software BoneXpert™.

RESULTS

There was a strong correlation between the automatic and the manual method (r = 0.983, p < 0.001). The automatic method tended to generate higher bone age values (0.64 ± 0.73 years) in the younger patients (4-5 years) and to underestimate retardation or acceleration of bone age. The so-called "bone health index" (BHI) was reduced in comparison to the reference population. Bone health index standard deviation score (BHI-SDS) was not related to the stage of CKD, but weakly negatively correlated with plasma PTH concentrations (r = 0.12, p = 0.019).

CONCLUSIONS

BoneXpert™ allows an objective, time-saving, and in general valid bone age assessment in children with CKD. Possible underestimation of retarded or accelerated bone age should be taken into account. Validation of the BHI needs further study.

Hand X-ray in pediatric endocrinology: Skeletal age assessment and beyond

Skeletal age assessment (SAA) is a clinical procedure which is used in determining the SA of children and adolescents. Bone development is influenced by a number of factors, including nutrition, hormonal secretions, and genetics. There are several factors to be borne in mind when using methods of assessing skeletal maturity. These include: Variability among methods, degree of variability in the estimation of skeletal maturation, sources of low accuracy, and dispersion of the values of skeletal maturation. Currently, the main clinical methods for SAA are the Greulich and Pyle (GP) and Tanner and Whitehouse (TW) methods. The GP method has the advantage of being quick and easy to use. A well-trained radiologist takes few minutes to determine the bone age (BA) from a single hand radiograph. The method of TW, however, seems to be more reliable than the GP method. In recent years, the increasing speed in computer sciences and reduction of their cost has given the opportunity to create and use computerized BA estimation system. Despite the fact that the number of automated systems for BAA have increased, most are still within the experimental phase. The use of automated BA determination system, cleared for clinical use in Europe (BoneXpert), has been validated for various ethnicities and children with endocrine disorders. Ultrasound imaging has some limitations that include operator dependence, lower intra-rater and inter-rater reliability of assessment and difficulties with standardization of documentation and imaging transfer. Magnetic resonance imaging (MRI) is noninvasive alternative tool for SA assessment in children. However, few studies have been reported on this topic, and further research is needed to evaluate the reliability and validity of MRI BAAs. In conclusion, at present radiographic methods for the assessment of BA remain the gold standards. Whatever method one adopts, it is essential to minimize the causes of imprecision by taking care to consider the quality of the X-ray. Moreover, it is imperative to assume a correct hand positioning because poor positioning can change the appearance of some bones. It is also preferable to employ scoring methods to these techniques and percentiles rather than BA in years and months. In addition, the possible differences in maturation among different population should be kept in mind.

Impact of ensemble learning in the assessment of skeletal maturity

The assessment of the bone age, or skeletal maturity, is an important task in pediatrics that measures the degree of maturation of children's bones. Nowadays, there is no standard clinical procedure for assessing bone age and the most widely used approaches are the Greulich and Pyle and the Tanner and Whitehouse methods. Computer methods have been proposed to automatize the process; however, there is a lack of exploration about how to combine the features of the different parts of the hand, and how to take advantage of ensemble techniques for this purpose. This paper presents a study where the use of ensemble techniques for improving bone age assessment is evaluated. A new computer method was developed that extracts descriptors for each joint of each finger, which are then combined using different ensemble schemes for obtaining a final bone age value. Three popular ensemble schemes are explored in this study: bagging, stacking and voting. Best results were achieved by bagging with a rule-based regression (M5P), scoring a mean absolute error of 10.16 months. Results show that ensemble techniques improve the prediction performance of most of the evaluated regression algorithms, always achieving best or comparable to best results. Therefore, the success of the ensemble methods allow us to conclude that their use may improve computer-based bone age assessment, offering a scalable option for utilizing multiple regions of interest and combining their output.

Automated bone age assessment: motivation, taxonomies, and challenges

Bone age assessment (BAA) of unknown people is one of the most important topics in clinical procedure for evaluation of biological maturity of children. BAA is performed usually by comparing an X-ray of left hand wrist with an atlas of known sample bones. Recently, BAA has gained remarkable ground from academia and medicine. Manual methods of BAA are time-consuming and prone to observer variability. This is a motivation for developing automated methods of BAA. However, there is considerable research on the automated assessment, much of which are still in the experimental stage. This survey provides taxonomy of automated BAA approaches and discusses the challenges. Finally, we present suggestions for future research.

Validation of automatic bone age determination in children with congenital adrenal hyperplasia

BACKGROUND

Determination of bone age is routinely used for following up substitution therapy in congenital adrenal hyperplasia (CAH) but today is a procedure with significant subjectivity.

OBJECTIVE

The aim was to test the performance of automatic bone age rating by the BoneXpert software package in all radiographs of children with CAH seen at our clinic from 1975 to 2006.

MATERIALS AND METHODS

Eight hundred and ninety-two left-hand radiographs from 100 children aged 0 to 17 years were presented to a human rater and BoneXpert for bone age rating. Images where ratings differed by more than 1.5 years were each rerated by four human raters.

RESULTS

Rerating was necessary in 20 images and the rerating result was closer to the BoneXpert result than to the original manual rating in 18/20 (90 %). Bone age rating precision based on the smoothness of longitudinal curves comprising a total of 327 data triplets spanning less than 1.7 years showed BoneXpert to be more precise (P<0.001).

CONCLUSION

BoneXpert performs reliable bone age ratings in children with CAH.

The reliability of the Greulich and Pyle atlas when applied to a modern Scottish population

This study examines the reliability of age estimation utilizing the Greulich and Pyle atlas in relation to a modern Scottish population. A total of 406 left-hand/wrist radiographs (157 females and 249 males) were age-assessed using the Greulich and Pyle atlas. Analysis showed that there was a strong correlation between chronological age and estimated age (females R(2) = 0.939, males R(2) = 0.940). When age groups were broken down into year cohorts, the atlas over-aged females from birth until 13 years of age. The pattern for males showed that the atlas under-estimated age until 13 years of age after which point it consistently over-aged boys between 13 and 17 years of age. This study showed that the Greulich and Pyle atlas can be applied to a modern population but would recommend that any analysis takes into account the potential for over- and under-aging shown in this study.

Radiographic analysis of epiphyseal fusion at knee joint to assess likelihood of having attained 18 years of age

Radiological analysis of the epiphyses of the knee joint provides new valuable information, which may be used in combination with these well-established techniques in order to maximise the accuracy in the assessment of age of 18 years. A total of 215 antero-posterior radiographs of the knee was reviewed retrospectively in patients aged between 14 and 24 years old (99 boys, 116 girls). Fusion was scored as stage 1, epiphysis not fused; stage 2, epiphysis is fully ossified and epiphyseal scar is visible; and stage 3, epiphysis is fully ossified and epiphyseal scar is not visible. Scores of 0, 1 and 2 were assigned to stages 1, 2 and 3, respectively. Lastly, the score related to epiphyseal fusion at the knee joint was obtained by adding the three scores of the distal femur, proximal tibia and proximal fibula. Age distribution gradually increased with each score, for both genders. The mean age (±standard error) in each score category varied between genders, but the differences were not significant (p > 0.11). Five tests were performed to discriminate between individuals who were or were not at age 18 years or more, according to the receiver operating curve. For boys, the highest value of accuracy was obtained with score 3, with high sensitivity (Se = 93.33 %) and specificity (Sp = 89.29 %). For girls, it was obtained with score 4, with high accuracy (Acc = 85.86 %). These results indicate that radiographic analysis of the knee is a valuable alternative as a non-invasive method of estimation of 18 years of age.

Correlation between dental maturation and chronological age in patients with cerebral palsy, mental retardation, and Down syndrome

Determining a child's chronological age and stage of maturation is particularly important in fields such as paediatrics, orthopaedics, and orthodontics, as well as in forensic and anthropological studies. Some systemic conditions can cause abnormal physiological maturation, and skeletal maturation is usually more delayed than dental maturation. The aim of this study was to determine dental age in a group of patients with the most prevalent congenital or perinatally occurring physical and mental disabilities. The study group comprised 155 white Spanish children aged 3-17 years (35 with cerebral palsy, 83 with mental retardation and no associated syndromes or systemic conditions, and 37 with Down syndrome). The dental maturation indices described by Nolla and Demirjian were used to generate regression lines for the dental age of individuals in a control group (688 white Spanish children aged 3-17 years) and the formulae were then used to determine the dental age of patients in the study group. No significant differences were found between dental and chronological age in boys with cerebral palsy, mental retardation, or Down syndrome. In contrast, dental age (calculated from the linear regression model that included values for the Demirjian index) was significantly delayed compared with chronological age in girls with cerebral palsy or Down syndrome.

Applicability of Greulich and Pyle method for age assessment in forensic practice on an Italian sample

BACKGROUND

The main importance in age estimation lies in the assessment of criminal liability and protection of unaccompanied minor immigrants, when their age is unknown. Under Italian law, persons are not criminally responsible before they reach the age of 14. The age of 18 is important when deciding whether juvenile or adult law must be applied. In the case of unaccompanied minors, it is important to assess age in order to establish special protective measures, and correct age estimation may prevent a person over 18 from benefiting from measures reserved for minors.

OBJECTIVE

Since the Greulich and Pyle method is one of the most frequently used in age estimation, the aim of this study was to assess the reproducibility and accuracy of the method on a large Italian sample of teenagers, to ascertain the applicability of the Atlas at the critical age thresholds of 14 and 18 years.

MATERIALS AND METHODS

This retrospective study examined posteroanterior X-ray projections of hand and wrist from 484 Italian-Caucasian young people (125 females, 359 males) between 11 and 19 years old. All radiographic images were taken from trauma patients hospitalized in the Azienda Ospedaliero Universitaria Ospedali Riuniti of Ancona (Italy) between 2006 and 2007. Two physicians analyzed all radiographic images separately. The blind method was used.

RESULTS

In the case of an estimated age of 14 years old, the true age ranged from 12.2 to 15.9 years (median, 14.3 years, interquartile range, 1.0 years) for males, and 12.6 to 15.7 years (median, 14.2 years, interquartile range, 1.7 years) for females. In the case of an estimated age of 18 years, the true age ranged from 15.6 to 19.7 years (median, 17.7 years, interquartile range, 1.4 years) for males, and from 16.2 to 20.0 years (median, 18.7 years, interquartile range, 1.8 years) for females.

CONCLUSION

Our study shows that although the GPM is a reproducible and repeatable method, there is a wide margin of error in the estimation of chronological age, mainly in the critical estimated ages of 14 and 18 years old in both males and females.

Automatic determination of Greulich and Pyle bone age in healthy Dutch children

BACKGROUND

Bone age (BA) assessment is a routine procedure in paediatric radiology, for which the Greulich and Pyle (GP) atlas is mostly used. There is rater variability, but the advent of automatic BA determination eliminates this.

OBJECTIVE

To validate the BoneXpert method for automatic determination of skeletal maturity of healthy children against manual GP BA ratings.

MATERIALS AND METHODS

Two observers determined GP BA with knowledge of the chronological age (CA). A total of 226 boys with a BA of 3-17 years and 179 girls with a BA of 3-15 years were included in the study. BoneXpert's estimate of GP BA was calibrated to agree on average with the manual ratings based on several studies, including the present study.

RESULTS

Seven subjects showed a deviation between manual and automatic BA in excess of 1.9 years. They were re-rated blindly by two raters. After correcting these seven ratings, the root mean square error between manual and automatic rating in the 405 subjects was 0.71 years (range 0.66-0.76 years, 95% CI). BoneXpert's GP BA is on average 0.28 and 0.20 years behind the CA for boys and girls, respectively.

CONCLUSION

BoneXpert is a robust method for automatic determination of BA.

Clinical application of automated Greulich-Pyle bone age determination in children with short stature

BACKGROUND

Bone age (BA) rating is time consuming and highly rater dependent.

OBJECTIVE

To adjust the fully automated BoneXpert method to agree with the manual Greulich and Pyle BA (GP BA) ratings of five raters and to validate the accuracy for short children.

MATERIALS AND METHODS

A total of 1,097 left hand radiographs from 188 children with short stature, including growth hormone deficiency (44%) and Turner syndrome (29%) were evaluated.

RESULTS

BoneXpert rejected 14 of the 1,097 radiographs, and deviated by more than 1.9 years from the operator BA for 27 radiographs. These were rerated blindly by four operators. Of the 27 new ratings, 26 were within 1.9 years of the automatic BA values. The root mean square deviation between manual and automatic rating was 0.72 years (95% CI 0.69-0.75).

CONCLUSION

BoneXpert's ability to process 99% of images automatically without errors, and to obtain good agreement with an operator suggests that the method is efficient and reliable for short children.

Biases in radiologic reasoning

OBJECTIVE

The purpose of this article is to outline common biases in medical reasoning that contribute to avoidable errors in diagnostic and therapeutic decision making.

CONCLUSION

By recognizing and understanding common biases in medical reasoning, we can more effectively counteract them.

The BoneXpert method for automated determination of skeletal maturity

Bone age rating is associated with a considerable variability from the human interpretation, and this is the motivation for presenting a new method for automated determination of bone age (skeletal maturity). The method, called BoneXpert, reconstructs, from radiographs of the hand, the borders of 15 bones automatically and then computes "intrinsic" bone ages for each of 13 bones (radius, ulna, and 11 short bones). Finally, it transforms the intrinsic bone ages into Greulich Pyle (GP) or Tanner Whitehouse (TW) bone age. The bone reconstruction method automatically rejects images with abnormal bone morphology or very poor image quality. From the methodological point of view, BoneXpert contains the following innovations: 1) a generative model (active appearance model) for the bone reconstruction; 2) the prediction of bone age from shape, intensity, and texture scores derived from principal component analysis; 3) the consensus bone age concept that defines bone age of each bone as the best estimate of the bone age of the other bones in the hand; 4) a common bone age model for males and females; and 5) the unified modelling of TW and GP bone age. BoneXpert is developed on 1559 images. It is validated on the Greulich Pyle atlas in the age range 2-17 years yielding an SD of 0.42 years [0.37; 0.47] 95% conf, and on 84 clinical TW-rated images yielding an SD of 0.80 years [0.68; 0.93] 95% conf. The precision of the GP bone age determination (its ability to yield the same result on a repeated radiograph) is inferred under suitable assumptions from six longitudinal series of radiographs. The result is an SD on a single determination of 0.17 years [0.13; 0.21] 95% conf.

Prostate Cancer Localization with Endorectal MR Imaging and MR Spectroscopic Imaging: Effect of Clinical Data on Reader Accuracy

PURPOSE

To determine the effect of digital rectal examination findings, sextant biopsy results, and prostate-specific antigen (PSA) levels on reader accuracy in the localization of prostate cancer with endorectal magnetic resonance (MR) imaging and MR spectroscopic imaging.

MATERIALS AND METHODS

This was a retrospective study of 37 patients (mean age, 57 years) with biopsy-proved prostate cancer. Transverse T1-weighted, transverse high-spatial-resolution, and coronal T2-weighted MR images and MR spectroscopic images were obtained. Two independent readers, unaware of clinical data, recorded the size and location of suspicious peripheral zone tumor nodules on a standardized diagram of the prostate. Readers also recorded their degree of diagnostic confidence for each nodule on a five-point scale. Both readers repeated this interpretation with knowledge of rectal examination findings, sextant biopsy results, and PSA level. Step-section histopathologic findings were the reference standard. Logistic regression analysis with generalized estimating equations was used to correlate tumor detection with clinical data, and alternative free-response receiver operating characteristic (AFROC) curve analysis was used to examine the overall effect of clinical data on all positive results.

RESULTS

Fifty-one peripheral zone tumor nodules were identified at histopathologic evaluation. Logistic regression analysis showed awareness of clinical data significantly improved tumor detection rate (P <.02) from 15 to 19 nodules for reader 1 and from 13 to 19 nodules for reader 2 (27%-37% overall) by using both size and location criteria. AFROC analysis showed no significant change in overall reader performance because there was an associated increase in the number of false-positive findings with awareness of clinical data, from 11 to 21 for reader 1 and from 16 to 25 for reader 2.

CONCLUSION

Awareness of clinical data significantly improves reader detection of prostate cancer nodules with endorectal MR imaging and MR spectroscopic imaging, but there is no overall change in reader accuracy, because of an associated increase in false-positive findings. A stricter definition of a true-positive result is associated with reduced sensitivity for prostate cancer nodule detection.