Data Augmentation of Neonatal Thermal Images Using Deep Learning

Karthik, K.G.V.S.; Nithin, K.; Dhanush, Bavisetti Sai; Praveen, K.; Sarath, S

doi:10.1109/icccnt51525.2021.9579769

Cited by 5 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A comparison of the proposed approach with related studies in the neonatal context [ 24 , 25 ] revealed that the amount of training data (without classical augmentation) was in the same range as our dataset. However, the datasets used in these studies (thermal images and MRI scans) were much more homogeneous.…”

Section: Discussionmentioning

confidence: 96%

“…Next to the application of neonatal MR image processing, the use of GANs to augment images showing infants was less covered in the literature. Very recently, Karthik et al used a DCGAN to augment a dataset of neonatal thermal images [ 25 ]. Although GANs have been used for various neonatal applications, so far, the augmentation of a neonatal RGB image dataset, which could be mentioned to precisely categorize this work, has not been described in the literature.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Conditional Generative Adversarial Networks for Data Augmentation of a Neonatal Image Dataset

Lyra

Mustafa

Rixen

et al. 2023

Sensors

View full text Add to dashboard Cite

In today’s neonatal intensive care units, monitoring vital signs such as heart rate and respiration is fundamental for neonatal care. However, the attached sensors and electrodes restrict movement and can cause medical-adhesive-related skin injuries due to the immature skin of preterm infants, which may lead to serious complications. Thus, unobtrusive camera-based monitoring techniques in combination with image processing algorithms based on deep learning have the potential to allow cable-free vital signs measurements. Since the accuracy of deep-learning-based methods depends on the amount of training data, proper validation of the algorithms is difficult due to the limited image data of neonates. In order to enlarge such datasets, this study investigates the application of a conditional generative adversarial network for data augmentation by using edge detection frames from neonates to create RGB images. Different edge detection algorithms were used to validate the input images’ effect on the adversarial network’s generator. The state-of-the-art network architecture Pix2PixHD was adapted, and several hyperparameters were optimized. The quality of the generated RGB images was evaluated using a Mechanical Turk-like multistage survey conducted by 30 volunteers and the FID score. In a fake-only stage, 23% of the images were categorized as real. A direct comparison of generated and real (manually augmented) images revealed that 28% of the fake data were evaluated as more realistic. An FID score of 103.82 was achieved. Therefore, the conducted study shows promising results for the training and application of conditional generative adversarial networks to augment highly limited neonatal image datasets.

show abstract

Section: Discussionmentioning

confidence: 96%

Section: Related Workmentioning

confidence: 99%