A novel approach for yoga pose estimation based on in-depth analysis of human body joint detection accuracy

Desai, Miral M.; Mewada, Hiren

doi:10.7717/peerj-cs.1152

Cited by 12 publications

(6 citation statements)

References 28 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ground truths dataset was prepared by segmenting the CBCT images (DICOM format) using a third-party A.I. tool (Makesense.AI) [ 32 ]. Teeth were grouped as 0 (ERR depth = 0.5 mm), 1 (ERR depth = 1.0 mm), 2 (ERR depth = 2.0 mm), 3 (no ERR).…”

Section: Methodsmentioning

confidence: 99%

Application of deep learning and feature selection technique on external root resorption identification on CBCT images

Reduwan,

Abdul Aziz,

Mohd Razi

et al. 2024

BMC Oral Health

View full text Add to dashboard Cite

Background Artificial intelligence has been proven to improve the identification of various maxillofacial lesions. The aim of the current study is two-fold: to assess the performance of four deep learning models (DLM) in external root resorption (ERR) identification and to assess the effect of combining feature selection technique (FST) with DLM on their ability in ERR identification. Methods External root resorption was simulated on 88 extracted premolar teeth using tungsten bur in different depths (0.5 mm, 1 mm, and 2 mm). All teeth were scanned using a Cone beam CT (Carestream Dental, Atlanta, GA). Afterward, a training (70%), validation (10%), and test (20%) dataset were established. The performance of four DLMs including Random Forest (RF) + Visual Geometry Group 16 (VGG), RF + EfficienNetB4 (EFNET), Support Vector Machine (SVM) + VGG, and SVM + EFNET) and four hybrid models (DLM + FST: (i) FS + RF + VGG, (ii) FS + RF + EFNET, (iii) FS + SVM + VGG and (iv) FS + SVM + EFNET) was compared. Five performance parameters were assessed: classification accuracy, F1-score, precision, specificity, and error rate. FST algorithms (Boruta and Recursive Feature Selection) were combined with the DLMs to assess their performance. Results RF + VGG exhibited the highest performance in identifying ERR, followed by the other tested models. Similarly, FST combined with RF + VGG outperformed other models with classification accuracy, F1-score, precision, and specificity of 81.9%, weighted accuracy of 83%, and area under the curve (AUC) of 96%. Kruskal Wallis test revealed a significant difference (p = 0.008) in the prediction accuracy among the eight DLMs. Conclusion In general, all DLMs have similar performance on ERR identification. However, the performance can be improved by combining FST with DLMs.

show abstract

Section: Methodsmentioning

confidence: 99%

Application of deep learning and feature selection technique on external root resorption identification on CBCT images

Reduwan,

Abdul Aziz,

Mohd Razi

et al. 2024

BMC Oral Health

View full text Add to dashboard Cite

show abstract

“…tool (Makesense.AI.) [31]. Teeth were grouped as 0 (ERR depth = 0.5 mm), 1 (ERR depth = 1.0 mm), 2 (ERR depth= 2.0 mm), 3 (no ERR).…”

Section: Ground Truth Labellingmentioning

confidence: 99%

Application of Deep Learning and Feature Selection Technique on External Root Resorption Identification on CBCT Images

Reduwan,

Aziz,

Razi

et al. 2023

Preprint

View full text Add to dashboard Cite

Background: Artificial intelligence have been proven to improve the identification of various maxillofacial lesions. The aim of the current study is two-fold, to assess the performance of four deep learning models (DLM) in external root resorption (ERR) identification, and to assess the effect of combining feature selection technique (FST) with DLM on their ability in ERR identification. Methods: External root resorption was simulated on 88 extracted premolar teeth using tungsten bur according to different depths (0.5mm, 1mm and 2mm). All teeth were scanned using a Cone beam CT (Carestream Dental-CHECK). A training (70%), validation (10%), and test (20%) dataset were established. The performance of four DLMs (i. Random Forest (RF)+Visual Geometry Group 16 (VGG), ii. RF+EfficienNetB4 (EFNET), iii. Support Vector Machine (SVM)+VGG and iv. SVM+EFNET) and four hybrid models (DLM+FST: i. FS+RF+VGG, ii. FS+RF+EFNET, iii. FS+SVM+VGG and iv. FS+SVM+EFNET) was compared. Five performance parameters were assessed namely classification accuracy, F1-score, precision, specificity, error rate. FST algorithms (Boruta and Recursive Feature Selection) were combined with the DLMs to assess their performance. Result: RF+VGG exhibited the highest performance in identifying ERR followed by the other tested models. Similarly, FST combined with RF+VGG outperformed other models with classification accuracy, F1-score, precision, and specificity of 81.9%, weighted accuracy of 83%, and AUC of 96%. Conclusion: In general, all DLMs have similar performance on ERR identification. However, the performance can be improved by combining FST with DLMs.

show abstract

“…Similarly, another study by Desai & Mewada (2023) developed an AR-based soccer training system that used CV and AI technology to track the user's movements and provide real-time feedback on their performance [19]. The system included various training scenarios and challenges that required the user to perform specific movements, enhancing their motivation and engagement in the training process.…”

Section: Related Workmentioning

confidence: 99%

Development of Computer Vision-enabled Augmented Reality Games to Increase Motivation for Sports

Doskarayev¹,

Omarov²,

Omarov³

et al. 2023

IJACSA

View full text Add to dashboard Cite

This research paper presents the development of computer vision-enabled augmented reality games based on action detection to increase motivation for sports. With the increasing popularity of digital games, physical activity and sports participation have been declining, especially among the younger generation. To address this issue, we developed a series of augmented reality games that require players to perform physical actions to progress and succeed in the game. These games were developed using computer vision technology to detect the players' movements and provide real-time feedback, enhancing the gaming experience and promoting physical activity. The results of our user study showed that participants who played the augmented reality games reported higher levels of motivation to engage in physical activity and sports. The findings suggest that computer vision-enabled augmented reality games can be an effective tool to promote physical activity and sports participation, especially among younger generations.

show abstract

A novel approach for yoga pose estimation based on in-depth analysis of human body joint detection accuracy

Cited by 12 publications

References 28 publications

Application of deep learning and feature selection technique on external root resorption identification on CBCT images

Application of deep learning and feature selection technique on external root resorption identification on CBCT images

Application of Deep Learning and Feature Selection Technique on External Root Resorption Identification on CBCT Images

Development of Computer Vision-enabled Augmented Reality Games to Increase Motivation for Sports

Contact Info

Product

Resources

About