Research Papers - Dept of Computer Systems Engineering

Permanent URI for this collection https://rda.sliit.lk/handle/123456789/1253

Browse

Filters

2020 - 20222

Advanced Search

Filter by

Settings

Subject: search.filters.subject.Intelligent Disease

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    PublicationEmbargo
    Multilingual Conversational AI incorporated with Visual Questions Answering and Intelligent Disease Prediction for Healthcare Industry
    (IEEE, 2022-07-18) Sasmitha, N. U. A.; Wathasha, H. K. G. V.; Guruge, P. P. L.; Silva, W. J. T.; Rupasinghe, L; Gunarathne, G. W. D. A.
    Artificial intelligence (AI) is becoming more active than ever in everyday life and steadily being incorporated to healthcare. AI, with its seemingly limitless power, affirms a promising future to a revolutionized healthcare system. This paper is proposing a conversational AI solution in two different languages, English and Sinhala, to predict diseases through a conversation, a visual question answering solution to generate answers are based on a given question and a medical image and a disease forecasting module. A robust, accurate prediction is a rather difficult task given the availability of data and absence of preprocessed, clean data. With the aid of outlier rejection, data imputation, vectorization, feature selection and data standardization, the proposed framework gets the advantage of latest machine learning advancements such as AI using DIET classifier and NLU pipelines, for the conversational disease diagnosis which uses support vector machine (SVM) achieved an accuracy of 0.93. Moreover, the visual questions answering module with VGG16 preprocessing, GoogleNews vectors, LSTM networks, scores an accuracy of 0.9721. In addition, time series analysis models such as ARIMA and adaptive models using PROPHET library for forecasting diseases, classification using random forest scoring an accuracy of 0.81, logistic regression scoring an accuracy of 0.84 for predicting diseases. The objective of this research is to compare and select the best fitting models to be used for a centralized framework for healthcare industry.
  • Thumbnail Image
    PublicationEmbargo
    Intelligent disease detection system for greenhouse with a robotic monitoring system
    (IEEE, 2020-12-10) Fernando, S; Nethmi, R; Silva, A; Perera, A; De Silva, R; Abeygunawardhana, P. K. W
    Greenhouse farming plays a significant role in the agricultural industry because of its controlled climatic features. Recent examinations have stated that the mean creation of the yields under greenhouses is lessening due to disease events in the plants. These foods have become an imposing undertaking because these plants are being assaulted by different bacterial diseases, micro-organisms, and pests. The chemicals are applied to the plants intermittently without thinking about the necessity of each plant. Several problems have occurred in the greenhouse environment due to these causes. Therefore, there is a huge necessity for a system to detect diseases at an early stage. This research focused on designing a system to detect disease, which causes yellowish in greenhouse plants. Plant yellowing can be considered a significant problem of plants that grow under greenhouse-controlled environments. Through this research is focused on the most important and one of the most attention-grabbing crop tomato. There are specific diseases that cause yellowish the tomato plant, and they have been identified. The techniques utilized for early recognition of infection are image processing, machine learning, and deep learning.