scholarly journals Artificial intelligence system can achieve comparable results to experts for bone age assessment of Chinese children with abnormal growth and development

PeerJ ◽  
2020 ◽  
Vol 8 ◽  
pp. e8854
Author(s):  
Fengdan Wang ◽  
Xiao Gu ◽  
Shi Chen ◽  
Yongliang Liu ◽  
Qing Shen ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Growth And Development ◽  
Bone Age ◽  
Chinese Children ◽  
Absolute Difference ◽  
Reference Standard ◽  
Limits Of Agreement ◽  
Altman Plot ◽  
Bland Altman Plot ◽  
Abnormal Growth

Objective Bone age (BA) is a crucial indicator for revealing the growth and development of children. This study tested the performance of a fully automated artificial intelligence (AI) system for BA assessment of Chinese children with abnormal growth and development. Materials and Methods A fully automated AI system based on the Greulich and Pyle (GP) method was developed for Chinese children by using 8,000 BA radiographs from five medical centers nationwide in China. Then, a total of 745 cases (360 boys and 385 girls) with abnormal growth and development from another tertiary medical center of north China were consecutively collected between January and October 2018 to test the system. The reference standard was defined as the result interpreted by two experienced reviewers (a radiologist with 10 years and an endocrinologist with 15 years of experience in BA reading) through consensus using the GP atlas. BA accuracy within 1 year, root mean square error (RMSE), mean absolute difference (MAD), and 95% limits of agreement according to the Bland-Altman plot were statistically calculated. Results For Chinese pediatric patients with abnormal growth and development, the accuracy of this new automated AI system within 1 year was 84.60% as compared to the reference standard, with the highest percentage of 89.45% in the 12- to 18-year group. The RMSE, MAD, and 95% limits of agreement of the AI system were 0.76 years, 0.58 years, and −1.547 to 1.428, respectively, according to the Bland-Altman plot. The largest difference between the AI and experts’ BA result was noted for patients of short stature with bone deformities, severe osteomalacia, or different rates of maturation of the carpals and phalanges. Conclusions The developed automated AI system could achieve comparable BA results to experienced reviewers for Chinese children with abnormal growth and development.

scholarly journals Correlation of Corneal Thickness Measurements Between Tono-pachymeter and Scheimpflug- Placido Topography

2021 ◽  
Vol 16 (2) ◽  
pp. 161-167
Author(s):  
Zeynep Eylül Ercan ◽  
Keyword(s):  
Corneal Thickness ◽  
Age Groups ◽  
Limits Of Agreement ◽  
Altman Plot ◽  
Bland Altman Plot ◽  
Significant Relationships ◽  
Plot Analysis ◽  
Group 2 ◽  
Thickness Measurements ◽  
Group 1

Central corneal thickness (CCT) measurements are important for diagnosis, treatment, and surgery planning in ophthalmology. The purpose of this study was to see whether CCT measurements taken with Tono-pachymeter and Scheimpflug- Placido Topography had any significant differences. Tono-pachymeter and topography CCT measurements were taken (n=400). Inter-measurement agreement between them was determined using Bland-Altman Plot analysis. Age groups were also formed as group 1 (aged 18-50 years, 94 males, 106 females) and group 2 (age >51 years, 100 males, 100 females). Mean CCTs measured by Tonopachymeter and topography were 563.77 +±26.43 and 560.88 + 26.341 microns. Bland-Altman Plot analysis showed in total, 13 were above the upper limit and 5 were under the minimum limit of agreement with regression analysis showing no significant relationships (p=0.213). Group 1 had 7 above and 2 below from the limits of agreement. Group two had 9 above and 2 below from the limits of agreement. Both groups showed insignificant differences between devices (p=0.07 and p=0.86). Tono-pachymeter and Scheimpflug-Placido Topography give reliable CCT results within each other. However, since the limit of agreement ranges can still affect one-to-one patient evaluations, we recommend clinics that use these devices to not interchange measurements in practice.

Evaluating the agreement between the empirical volumes of tumor lesions and the estimated volumes of tumor lesions based on the longest diameter.

2012 ◽  
Vol 30 (4_suppl) ◽  
pp. 635-635 ◽  
Author(s):  
Ruediger P. Laubender ◽  
Michael Schlichting ◽  
Ute Sartorius ◽  
Dominik Paul Modest ◽  
Ulrich Robert Mansmann ◽  
...  
Keyword(s):  
Tumor Volume ◽  
Three Dimensional ◽  
Disease Process ◽  
Volume Data ◽  
Limits Of Agreement ◽  
Altman Plot ◽  
Bland Altman Plot ◽  
Volume Measurements ◽  
Colorectal Cancer Patients ◽  
The Impact

635 Background: Tumor lesions are commonly evaluated by RECIST or WHO criteria, which involves reducing the development of a three-dimensional tumor to a two-dimensional process. However, volume measurements might reflect the clinical impact of the disease process more precisely. Therefore, we investigated if the volume of a tumor lesion can be reconstructed from RECIST- or WHO-based measurements. Methods: 20 colorectal cancer patients underwent routine staging MDCT examinations as part of the multicenter phase II CIOX trial comparing cetuximab + XELOX versus cetuximab + XELIRI. Two hepatic target lesions per patient were defined at baseline and measured using 1) semi-automated volumetry (Siemens Syngo Via Oncology, Siemens Healthcare) with manual correction and 2) semi-automated measurement of the longest diameter (LD) according to RECIST 1.1 and the corresponding longest orthogonal diameter (LOD). Patient lesions were followed over time using the same measurement strategies. An algorithm was developed for reconstructing the volume of a lesion based on the LD and the corresponding LOD. Agreement of the volume data generated by both methods was analyzed by a Bland-Altman plot. The limits of agreements were calculated using a variance components model considering repeated tumor assessments based on the log-transformed sum over the volume of a patient’s lesions. Results: 151 lesions from the 20 patients were measured at 73 tumor assessments. Volume was slightly overestimated by the algorithm compared to the volumetric measurements (p = 0.07). The Bland-Altman plot showed good agreement (mean difference −0.05, limits of agreement [−0.40, 0.30]). As expected, three out of the 73 measurements fell outside the limits of agreement. Conclusions: The proposed algorithm for the WHO-based reconstruction of the tumor volume provides a good approximation to the true sum of the volumes of the lesions. In future studies, volumetric tumor information can be included even if not all centers have full volumetric software available. This enables further research on the impact of changes in tumor volume on disease prognosis and/or prediction of treatment efficacy.

scholarly journals Correlation and agreement between superb micro-vascular imaging and contrast-enhanced ultrasound for assessing radiofrequency ablation treatment of thyroid nodules: a preliminary study

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Yu Lan ◽  
Nan Li ◽  
Qing Song ◽  
Ming-bo Zhang ◽  
Yu-kun Luo ◽  
...  
Keyword(s):  
Volume Measurement ◽  
Vascular Imaging ◽  
Scatter Plot ◽  
Limits Of Agreement ◽  
Altman Plot ◽  
Bland Altman Plot ◽  
Observer Reliability ◽  
Contrast Enhanced ◽  
Good Agreement

Abstract Background To evaluate the correlation and agreement between superb micro-vascular imaging (SMI) mode and the contrast-enhanced ultrasound (CEUS) mode for the ablative completeness and the volumes of ablation lesions to determine the clinical application value of SMI in follow-up after radiofrequency ablation. Methods From April 2020 to June 2020, two radiologists used SMI and CEUS mode to measure the volume of the ablation lesion. We use intra-class correlation coefficient (ICC), scatter plots and Bland–Altman plots to evaluate the correlation and agreement of the two techniques. In addition, intra- and inter-observer reliability in volume measurement of ablation lesions with SMI mode was assessed. Results SMI mode and CEUS mode have good agreement in the evaluation of ablative completeness. The ICC was 0.876 and 0.928 of reader A and reader B between SMI mode and CEUS mode in terms of ablation lesions volume measurement. There was a strong correlation between the two modes in both reader A and reader B (rA = 0.808; rB = 0.882). The ICC was 0.836 for the inter-observer reliability of SMI technique. The scatter plot showed a good linear relation (r = 0.715). In the Bland–Altman plot, 4.35% (1/23) of the points was outside the 95% limits of agreement. The ICC was 0.965 for the intra-observer reliability of SMI technique, the scatter plot also showed a strong linear correlation (r = 0.965). In the Bland–Altman plot, 8.70% (2/23) of the points was outside the 95% limits of agreement. Conclusions SMI and CEUS have good agreement and correlation in the ablation volume measurement. SMI technology is expected to be applied as an alternative to CEUS in the clinical follow-up of ablation lesions.

Performance of an artificial intelligence system for bone age assessment in Tibet

2021 ◽  
Vol 94 (1120) ◽  
pp. 20201119
Author(s):  
Fengdan Wang ◽  
Wangjiu Cidan ◽  
Xiao Gu ◽  
Shi Chen ◽  
Wu Yin ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Southern China ◽  
Medical Center ◽  
Bone Age ◽  
Absolute Difference ◽  
Bone Maturation ◽  
Mean Absolute Difference ◽  
Left Hand ◽  
Intelligence System ◽  
Bone Age Assessment

Objective: To investigate whether bone age (BA) of children living in Tibet Highland could be accurately assessed using a fully automated artificial intelligence (AI) system. Methods： Left hand radiographs of 385 children (300 Tibetan and 85 immigrant Han) aged 4–18 years who presented to the largest medical center of Tibet between September 2013 and November 2019 were consecutively collected. From these radiographs, BA was determined using the Greulich and Pyle (GP) method by experts in a consensus manner; furthermore, BA was estimated by a previously reported artificial intelligence (AI) BA system based on Han children from southern China. The performance of the AI system was compared with that of experts by using statistical analysis. Results: Compared with the experts’ results, the accuracy of the AI system for Tibetan and Han children within 1 year was 84.67 and 89.41%, respectively, and its mean absolute difference (MAD) was 0.65 and 0.56 years, respectively. The discrepancy in hand-wrist bone maturation was the main cause of low accuracy of the system in the 4- to 6-year-old group. Conclusion: The AI BA system developed for Han Chinese children living in flat regions could enable to assess BA accurately in Tibet where medical resources are limited. Advances in knowledge: AI-based BA system may serve as an effective and efficient solution to assess BA in Tibet.

scholarly journals Automated echocardiographic left ventricular strain measurements using deep learning

2021 ◽  
Vol 22 (Supplement_1) ◽  
Author(s):  
M Rogstadkjernet ◽  
SZ Zha ◽  
LG Klaeboe ◽  
CK Larsen ◽  
JM Aalen ◽  
...  
Keyword(s):  
Deep Learning ◽  
Left Ventricular ◽  
University Hospital ◽  
Absolute Difference ◽  
Lv Function ◽  
Altman Plot ◽  
Final Model ◽  
Strain Measurements ◽  
Bland Altman Plot ◽  
Model Training

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Research Council of Norway Background Speckle tracking echocardiography (STE) provides quantification of left ventricular (LV) deformation and is useful in the assessment of LV function. STE is increasingly being used clinically, and every effort to simplify and standardise STE is important. Manual outlining of regions of interest (ROIs) is labour intensive and may influence assessment of strain values. Purpose We hypothesised that ROIs from clinical echocardiographic exams could be used to train a deep learning (DL) model which could automate strain calculation by predicting ROIs with comparable fidelity to trained cardiologists. Methods Data consisted of still frames with cardiologist-defined ROIs from 435 clinical echocardiographic exams from a university hospital outpatient clinic. Exams included healthy subjects and patients with ischemic heart disease, heart failure, valvular disease and conduction abnormalities. Three frames surrounding mid-systole from standard apical views were paired with the mid-systolic ROI for training. Image quality was classified as high, medium or low by an experienced cardiologist prior to model training, and low quality images were excluded. The dataset was randomly split into 68%/17%/15% sets for training, validation and testing. A standard Unet architecture was used, with a combination of binary cross entropy and Dice loss functions. A total of 1025 images were used for final model training and testing, with augmentation used to extend the dataset. Predicted ROIs were reintroduced into commercially available echocardiographic analysis software to automatically calculate longitudinal strain (LS) values. Results The average absolute difference between manually measured and DL-assisted LS was 0.98 percentage points (95% confidence interval 0.82-1.13) for the final model. A Bland-Altman plot revealed very few outliers and no bias in the difference between manual and DL-assisted strain measurements. Conclusion The current study demonstrates that DL-assisted, automated strain measurement is feasible, and provides strain results within an acceptable range of intra- and interobserver variation. Employing DL in echocardiographic analysis could further facilitate adoption of STE parameters in clinical practice and research, and improve reproducibility. Abstract Figure. Bland-Altman plot. Manual and DL ROIs

scholarly journals Validation of a pre-coded food diary with energy expenditure, comparison of under-reporters v. acceptable reporters

2005 ◽  
Vol 94 (6) ◽  
pp. 998-1003 ◽  
Author(s):  
I. T. L. Lillegaard ◽  
L. F. Andersen
Keyword(s):  
Food Intake ◽  
Energy Expenditure ◽  
Pearson Correlation ◽  
Pearson Correlation Coefficient ◽  
Confidence Limits ◽  
Food Diary ◽  
Limits Of Agreement ◽  
Altman Plot ◽  
Bland Altman Plot ◽  
Males And Females

The objective of the present study was to compare energy intake (EI) assessed from a pre-coded food diary (PFD) with energy expenditure (EE) measured by a validated position-and-movement monitor (ActiReg®; PreMed AS, Oslo, Norway) in a group of Norwegian 9-year-olds. Moreover, we examined whether and how under-reporters (UR), identified with ActiReg®, differed from acceptable reporters (AR) according to food intake and BMI. A total of fifty-one 9-year-olds completed PFD and ActiReg®. The present study showed that on average EI was underestimated by 18 % compared with EE measured by ActiReg®. The 95 % confidence limits of agreement in a Bland–Altman plot for EI and EE varied from 1·97 MJ to −4·23 MJ (sd 2) among the girls and from 0·74 MJ to −5·26 MJ (sd 2) among the boys. The Pearson correlation coefficient between EI and EE was 0·28 (P=0·05) for males and females combined. Fifty-seven per cent of the participants were classified as AR, 39 % as UR and 4 % as over-reporters with the PFD. Under-reporting of energy remains a problem with the PFD method used in a group of 9-year-olds, especially among boys. However, UR and AR did not show a systematic misreporting related to macronutrients, unhealthy foods or BMI.

scholarly journals Design and Reliability of a Novel Heel Rise Test Measuring Device for Plantarflexion Endurance

2014 ◽  
Vol 2014 ◽  
pp. 1-7 ◽  
Author(s):  
Amy D. Sman ◽  
Claire E. Hiller ◽  
Adam Imer ◽  
Aldrin Ocsing ◽  
Joshua Burns ◽  
...  
Keyword(s):  
Interrater Reliability ◽  
Test Device ◽  
Limits Of Agreement ◽  
Altman Plot ◽  
Measuring Device ◽  
Bland Altman Plot ◽  
Measurement Protocol ◽  
Combined Action ◽  
And Performance ◽  
Gastrocnemius Muscles

Background. Plantarflexion results from the combined action of the soleus and gastrocnemius muscles in the calf. The heel rise test is commonly used to test calf muscle endurance, function, and performance by a wide variety of professionals; however, no uniform description of the test is available. This paper aims to document the construction and reliability of a novel heel rise test device and measurement protocol that is suitable for the needs of most individuals.Methods. This device was constructed from compact and lightweight materials and is fully adjustable, enabling the testing of a wide variety of individuals. It is easy to assemble and disassemble, ensuring that it is portable for use in different settings.Findings. We tested reliability on 40 participants, finding excellent interrater reliability (ICC2,10.97, 95% CI: 0.94 to 0.98). Limits of agreement were less than two repetitions in 90% of cases and the Bland-Altman plot showed no bias.Interpretation. We have designed a novel, standardized, simple, and reliable device and measurement protocol for the heel rise test which can be used by researchers and clinicians in a variety of settings.

Sign in / Sign up

Export Citation Format

Share Document