Publications by authors named "Prakash Kumar Karn"
Precision Segmentation of Subretinal Fluids in OCT Using Multiscale Attention-Based U-Net Architecture.
Bioengineering (Basel)
October 2024
Article Synopsis
- The paper introduces a deep-learning architecture aimed at segmenting retinal fluids in patients with Diabetic Macular Oedema (DME) and Age-related Macular Degeneration (AMD), addressing challenges in accuracy faced by existing techniques.
- The proposed model features an encoder-decoder network inspired by U-Net, utilizing enhanced OCT images with edge maps and advanced components like Residual and Inception modules alongside a multiscale attention mechanism for improved feature extraction.
- Results show the network achieves high F1 Scores on multiple datasets, indicating enhanced segmentation accuracy and clinical potential for diagnosing and managing retinal diseases.
Article Synopsis
- The paper introduces a new U-Net model featuring a hybrid attention mechanism designed to automate the segmentation of sub-retinal layers in Optical Coherence Tomography (OCT) images, addressing the challenges of manual segmentation.
- By integrating edge and spatial attention into the U-Net architecture, the model improves segmentation accuracy and focuses on important image features, promising to streamline the work for medical professionals.
- Evaluation results indicate high performance, with metrics such as a Dice score of 94.99% and an ARI of 97.00%, suggesting that this model is highly effective and could significantly enhance ocular imaging practices.
On Machine Learning in Clinical Interpretation of Retinal Diseases Using OCT Images.
Bioengineering (Basel)
March 2023
Optical coherence tomography (OCT) is a noninvasive imaging technique that provides high-resolution cross-sectional retina images, enabling ophthalmologists to gather crucial information for diagnosing various retinal diseases. Despite its benefits, manual analysis of OCT images is time-consuming and heavily dependent on the personal experience of the analyst. This paper focuses on using machine learning to analyse OCT images in the clinical interpretation of retinal diseases.
View Article and Find Full Text PDF