A DEEP LEARNING APPROACH LUNG SEGMENTATION AND PNEUMONIA DETECTION FROM X-RAYS

Lung segmentation is a process of detection and identification of lung cancer and pneumonia with the help of image processing techniques. Deep learning algorithms can be incorporated to build the computer-aided diagnosis (CAD) system for detecting or recognizing broad objects like acute respiratory distress syndrome (ARDS), Tuberculosis, Pneumonia, Lung cancer, Covid, and several other respiratory diseases. This paper presents pneumonia detection from lung segmentation using deep learning methods on chest radiography. Chest X-ray is the most useful technique among other existing techniques, due to its lesser cost. The main drawback of a chest x-ray is that it cannot detect all problems in the chest. Thus, implementing convolutional neural networks (CNN) to perform lung segmentation and to obtain correct results. The 'lost' regions of the lungs are reconstructed by an automatic segmentation method from raw images of chest X-ray. © 2022 IEEE.

Performance Analysis of GAN Approaches in the Portable Chest X-Ray Synthetic Image Generation for COVID-19 Screening

COVID-19 mainly affects lung tissues, aspect that makes chest X-ray imaging useful to visualize this damage. In the context of the global pandemic, portable devices are advantageous for the daily practice. Furthermore, Computer-aided Diagnosis systems developed with Deep Learning algorithms can support the clinicians while making decisions. However, data scarcity is an issue that hinders this process. Thus, in this work, we propose the performance analysis of 3 different state-of-the-art Generative Adversarial Networks (GAN) approaches that are used for synthetic image generation to improve the task of automatic COVID-19 screening using chest X-ray images provided by portable devices. Particularly, the results demonstrate a significant improvement in terms of accuracy, that raises 5.28% using the images generated by the best image translation model. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

DeepChestNet: Artificial intelligence approach for COVID-19 detection on computed tomography images

The conventional approach for identifying ground glass opacities (GGO) in medical imaging is to use a convolutional neural network (CNN), a subset of artificial intelligence, which provides promising performance in COVID-19 detection. However, CNN is still limited in capturing structured relationships of GGO as the texture and shape of the GGO can be confused with other structures in the image. In this paper, a novel framework called DeepChestNet is proposed that leverages structured relationships by jointly performing segmentation and classification on the lung, pulmonary lobe, and GGO, leading to enhanced detection of COVID-19 with findings. The performance of DeepChestNet in terms of dice similarity coefficient is 99.35%, 99.73%, and 97.89% for the lung, pulmonary lobe, and GGO segmentation, respectively. The experimental investigations on DeepChestNet-Lung, DeepChestNet-Lobe and DeepChestNet-COVID datasets, and comparison with several state-of-the-art approaches reveal the great potential of DeepChestNet for diagnosis of COVID-19 disease. © 2023 Wiley Periodicals LLC.

Processing of Wisconsin Breast Cancer Data using Ebola Optimization Algorithm with Mixture Kernel SVM

Being a deadly disease, breast cancer is becoming the more progressive one in providing higher mortality for females around the world. Thereby, the need for an appropriate strategy is always required for earlier breast cancer diagnosis. The physicians utilize the Computer-Aided Diagnosis (CAD) tool for effective and tireless detection of such cancers. In this regard, the work is intended to design a CAD system for breast cancer diagnosis in a timely manner. The implementation starts with the use of Wisconsin Breast Cancer dataset. After performing preprocessing and visual analysis of the input dataset, feature selection is performed to improve the efficiency of the CAD system. This can be done by using the recently evolved Ebola Optimization Algorithm (EOA). This algorithm is based on an effective approach used in the propagation of the Ebola virus among individuals. After feature selection, the dominant features are then classified with the aid of a mixture Kernel Support Vector Machine (mK-SVM) algorithm. Additionally, the work utilized the Linear SVM, and KNN algorithms for the experimental analysis and comparison. As a result, the mK-SVM together with EOA provides maximum accuracy of 97.19% in classifying the input as either benign severity or malignant case. © 2022 IEEE.

Automated classification method of COVID-19 cases from chest CT volumes using 2D and 3D hybrid CNN for anisotropic volumes

This paper proposes an automated classification method of chest CT volumes based on likelihood of COVID-19 cases. Novel coronavirus disease 2019 (COVID-19) spreads over the world, causing a large number of infected patients and deaths. Sudden increase in the number of COVID-19 patients causes a manpower shortage in medical institutions. Computer-aided diagnosis (CAD) system provides quick and quantitative diagnosis results. CAD system for COVID-19 enables efficient diagnosis workflow and contributes to reduce such manpower shortage. This paper proposes an automated classification method of chest CT volumes for COVID-19 diagnosis assistance. We propose a COVID-19 classification convolutional neural network (CNN) that has a 2D/3D hybrid feature extraction flows. The 2D/3D hybrid feature extraction flows are designed to effectively extract image features from anisotropic volumes such as chest CT volumes for diagnosis. The flows extract image features on three mutually perpendicular planes in CT volumes and then combine the features to perform classification. Classification accuracy of the proposed method was evaluated using a dataset that contains 1288 CT volumes. An averaged classification accuracy was 83.3%. The accuracy was higher than that of a classification CNN which does not have 2D and 3D hybrid feature extraction flows. © 2022 SPIE.

Radiographic Image Processing Analysis for Lung Infection - A Review

This paper presents a brief review on the developments of computer aided diagnosis system using image processing approaches. The rapid increase in lung infections which was in multiple during the current Corona virus infection has outcome with the need of automation system for an early detection of lung infection. Early detection of lung infection can avoid the spread of infection further and also act as an alarming intimation under critical cases. The need of such system has outcome with many researches in recent past towards developing new approaches toward improving the decision accuracy to reducing the system response time. This article review the past developments made in the area of developing automation systems with an analysis of attainted accuracy and methodology of image processing and classification system for automated lung infection detection. © 2022 IEEE.

Computer-aided diagnosis system for Covid-19 by using deep transfer learning through X-ray images

COVID-19 pandemic is triggering a massive epidemic in more than 180 countries worldwide, causing chaos in many people's health and lives. Identifying infected patients early enough and placing them under special treatment is one of the most critical steps in combating COVID-19. RT-PCR is a standard test process. The test procedure is typically conducted by air samples collected using a nasopharyngeal swab. However, using a nasal swab or sputum extract is not always possible. Due to the shortage of testing kits, virus mutations, and a longer time to detect. In addition to laboratory tests, chest scans can help diagnose COVID-19 in people who have severe clinical concerns. So, classification through X-ray images could be beneficial. This experiment aims to analyze the X-Ray images as abnormal or not. The intention is to train a convolution neural network(CNN) to classify the image using different architectures such as Xception, Resnet-50, DenseNet-121, VGG-16. Test the Performance metrics for each model and train further based on the insight gained. The following is an experimental study where we repeatedly train better models based on the insights gained from the previous model. The models tested on test data, and most of the results achieved a sensitivity rate of 98 percent (± 2 %), With a specificity rate of around 98 percent. While the achieved results are auspicious, additional research in a broader collection of COVID-19 chest X-ray pictures is needed to estimate accuracy rates accurately. © 2021 IEEE.

Segmentation of covid-19 lesions in CT images

The worldwide pandemic caused by the new coronavirus (COVID-19) has encouraged the development of multiple computer-aided diagnosis systems to automate daily clinical tasks, such as abnormality detection and classification. Among these tasks, the segmentation of COVID lesions is of high interest to the scientific community, enabling further lesion characterization. Automating the segmentation process can be a useful strategy to provide a fast and accurate second opinion to the physicians, and thus increase the reliability of the diagnosis and disease stratification. The current work explores a CNN-based approach to segment multiple COVID lesions. It includes the implementation of a U-Net structure with a ResNet34 encoder able to deal with the highly imbalanced nature of the problem, as well as the great variability of the COVID lesions, namely in terms of size, shape, and quantity. This approach yields a Dice score of 64.1%, when evaluated on the publicly available COVID-19-20 Lung CT Lesion Segmentation GrandChallenge data set. © 2021 IEEE

Covid19 Infection Detection and Classification Using CNN on Chest X-ray Images

Covid-19 has opened up a plethora of worries to the world since the past 2 years. The infection rate and death rate are increasing rapidly. It has worsened by the number of genetic mutations this virus has undergone. Timely detection of the disease is the only way out to handle this health emergency. Severity of this disease is when the virus attacks the major volume of the lung and results in pneumonia. To diagnose the pneumonia the first preferred modality is chest X-ray. There are two solid reasons why the Computer Aided Diagnosis (CAD) system is the need of the hour. First, the volume of X-rays generated for a huge number of infected patients to be assessed and second being the requirement of accuracy in diagnosis. Radiologists find it difficult to assess the severity through bare eyes and most of the time end up making a wrong conclusion which is chaotic decision. With the advent of technology, deep learning algorithms are proving to be most appropriate because of its ability to deliver expected accuracy and capacity to handle huge volume of data. This paper proposed a Deep Learning based Computer Aided Diagnosis System that accepts Chest X-ray image of a patient as input and classifies them as pneumonia or non-pneumonia. The Deep learning model is built and is trained with over 5000 chest X-ray images. Thus, trained model is then tested and validated and an accuracy of 96.66% is achieved. However, since the data is not real time, this work does not claim medical accuracy. The validation plots of the training loss and accuracy and validation loss and accuracy have been validated through regression. © 2021 IEEE.