Browsing by Author "Lanka, S."

Now showing 1 - 4 of 4

Ensemble neural models for ICD code prediction using unstructured and structured healthcare data
(Elsevier Ltd, 2024) Merchant, A.M.; Shenoy, N.; Lanka, S.; Kamath S․, S.
Disease coding is the process of assigning one or more standardized diagnostic codes to clinical notes that are maintained by health practitioners (e.g. clinicians) to track patient condition. Such a coding process is often expensive and error-prone, as human medical coders primarily perform it. Automating diagnostic coding using Artificial Intelligence is seen as an essential solution in Hospital Information Management Systems and approaches built on Convolutional Neural Networks currently perform best. In this work, a neural model built on unstructured clinical text for enabling automatic diagnostic coding for given patient discharge summaries is proposed. We incorporate a structured self-attention mechanism designed to boost learning of label-specific vectors and the significant clinical text snippets associated with a certain label for this purpose. These vectors are then combined with a novel code description pipeline leveraging the descriptions provided for each standardized diagnostic code. The proposed model achieved best performance in terms of standard metrics over the MIMIC-III dataset, outperforming models based on Longformers and Knowledge graphs. Furthermore, to leverage structured clinical data to enhance the proposed model, and to enable improved diagnostic code prediction, model ensembling is considered. A neural model built on structured data by leveraging supervised machine learning algorithms such as random forest and boosting, is designed for multi-class code classification. Experimental results revealed that the proposed ensemble models show promising performance compared to traditional models that rely solely on unstructured or structured clinical data, emphasizing their suitability for real-world deployment. © 2024 The Author(s)
Image Augmentation Strategies to Train GANs with Limited Data
(Springer Science and Business Media Deutschland GmbH, 2023) Lanka, S.; Velingkar, G.; Varadarajan, R.; Anand Kumar, M.
Training modern generative adversarial networks (GANs) to produce high-quality images requires massive datasets, which are challenging to obtain in many real-world scenarios, like healthcare. Training GANs on a limited dataset overfits the discriminator on the data to the extent that it cannot correctly distinguish between real and fake images. This paper proposes an augmentation mechanism to improve the datasetâ€™s size, quality, and diversity using a set of different augmentations, namely flipping of images, rotations, shear, affine transformations, translations, and a combination of these to form some hybrid augmentation. Fretchet distance has been used as the evaluation metric to analyze the performance of different augmentations on the dataset. It is observed that as the number of augmentations increase, the quality of generated images improves, and the Fretchet distance reduces. The proposed augmentations successfully improve the quality of generated images by the GAN when trained with limited data. Â© 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
Movie Box-Office Success Prediction Using Machine Learning
(Institute of Electrical and Electronics Engineers Inc., 2022) Velingkar, G.; Varadarajan, R.; Lanka, S.; Anand Kumar, M.
Being a multi-billion dollar business, the film industry contributes largely to helping sustain a country's economy. A movie's box office (the revenue generated by the number of tickets sold of a movie) is an essential indicator of the movie's popularity. It varies depending upon several factors, including a production company, genre, budget, reviews, ratings, etc. Predicting an approximate value for a movie's box office based upon the various parameters helps investors with this business make intelligent and informed decisions. Thus, this paper designs a machine learning model that can predict the revenue a film will generate based on the information available before the movie's release. It also provides a model capable of taking in the planned genre, the required revenue, and using the Random Forest Regression model, provides recommended budget, runtime, star power, and expected popularity. Â© 2022 IEEE.
Task Scheduling Using Deep Q-Learning
(Springer Science and Business Media Deutschland GmbH, 2022) Velingkar, G.; Kumar, J.K.; Varadarajan, R.; Lanka, S.; Anand Kumar, A.M.
Process scheduling is a very crucial task of operating systems. Effective scheduling ensures system efficiency and minimizes wastage of resources and cost overall, enhancing productivity. Most commonly, it is an exhaustive task to select the most accurate resources in executing these tasks. The solution for this effective job scheduling and resource management would preferably be dependent on the nature of the workload and adapt to any given environment compared to an algorithmic one. Thus, to meet this rising demand for an automated, self-assigning system, a deep Q-learning (Reinforcement learning technique)-based implementation has been done, which schedules tasks to maximize CPU utilization and memory utilization. Â© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.