• Parthasarathy

Pneumonia is a life-threatening lung infection so it requires initial recognition and exact classification in order to simplify quick medical intervention. This research addresses high-risk challenge of pneumonia analysis as well as classification by connecting highly-advanced technologies in medical imaging and machine learning (ML). This paper develops the Moth Flame Optimization with Long Short-Term Memory for Pneumonia Detection and Classification (MFOLSTM-PDC) model on Chest X-rays (CXR). Primary, we use a Wiener filter to improve the excellence of CXR images, decreasing noise and enhancing prominence of serious facts. And then, we remove high-level features from pre-processed images employing a DenseNet method. The detection stage employs an LSTM system which is a sort of recurrent neural network (RNN) that is suitable for consecutive data analysis. The LSTM is mainly trained in order to identify patterns and relations within removed features and then differentiate between pneumonia and non-pneumonia cases. We use MFO for parameter tuning to further improve the performance of the detection method. MFO is a nature-inspired optimization technique that perfects hyper-parameters of the LSTM method that guarantee optimum performance. The developed technique is verified on a huge dataset of CXR images by validating its efficiency in pneumonia recognition and classification. This complete procedure signifies a promising method for improvement of strong and exact pneumonia analysis systems with probable applications in medical settings to help healthcare experts in making decisions on time.

Chest X-ray; Moth flame optimization; Deep learning; Long short-term memory; Parameter tuning

"Moth Flame Optimization with Long Short-Term Memory for Pneumonia Detection and Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 10, page no.g291-g300, October-2023, Available :http://www.jetir.org/papers/JETIR2310533.pdf

"Moth Flame Optimization with Long Short-Term Memory for Pneumonia Detection and Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 10, page no. ppg291-g300, October-2023, Available at : http://www.jetir.org/papers/JETIR2310533.pdf