China
Africa
Explore chapters and articles related to this topic
Fractal Analysis in Histology Classification of Non-Small Cell Lung Cancer
Published in K.C. Santosh, Sameer Antani, D.S. Guru, Nilanjan Dey, Medical Imaging, 2019
Ravindra Patil, Geetha M. Srinidhi Bhat, Dinesh M.S. Leonard Wee, Andre Dekker
In both sexes combined, lung cancer is the most commonly diagnosed cancer (11.6% of the total cases) and the leading cause of cancer death. The total number of lung cancer cases in 2018 alone amounted to 2,093,876, the number of deaths with lung cancer being 1,761,007. Non-small cell lung cancer (NSCLC) accounts for 85% of all the lung cancers [1]. The cause of illness and the survival of NSCLC subjects vary across age, genetic profile, size of tumor, and histopathology of tumor. There are various studies that have established a correlation between the subtypes of NSCLC (squamous cell carcinoma, large cell carcinoma, adenocarcinoma, and “not otherwise specified”) to the patient’s survival. Also, it was studied that the prognosis for adenocarcinoma is poor compared to those for non-adenocarcinoma [2]. It was also concluded that surgical management should be different for each sub-category of NSCLC [3]. The current approach of subtype detection is performed using a biopsy procedure, where the tissue under observation is biopsied to determine the subtype, which is invasive in nature. The invasive approach is painful, costly, and not devoid of complications [4]. In recent times, several studies have been undertaken to identify the sub-categories of NSCLC non-invasively using radiomics, wherein large amount of quantitative features are mined and decision support models are built to achieve the desired objective [5]. Lately, radiomics has been applied to several medical problems such as tumors of lung, breast, and prostate, and also to images extracted from different medical imaging techniques (computed tomography (CT), magnetic resonance (MR), and positron emission tomography (PET)) [6–9, 10–13], showing promising results in each case.
A K-Means-Galactic Swarm Optimization-Based Clustering Algorithm with Otsu’s Entropy for Brain Tumor Detection
Published in Applied Artificial Intelligence, 2019
Satyasai Jagannath Nanda, Ishank Gulati, Rajat Chauhan, Rahul Modi, Uttam Dhaked
The proposed algorithms are tested on T2 weighted images with five types of brain tumor obtained from The Whole Brain Atlas Johnson and Becker (1999) Glioma: It is a kind of tumor that is mainly localized in brain and caused by glial cells. The tumor is generated due to three types of normal glial cells namely astrocytes, oligodendrocytes, and ependymal cells. Two techniques detect glioma-type tumors. Glioma fdg-PET: Glioma fdg-PET is an imaging technique for the detection of brain tumors. FDG(Fluoro-D-glucose) is a chemical that reacts with biologically active molecules inside the body and thus produces positron-radionuclide. The gamma rays are emitted by the positron-emitting radionuclide(tracer) PET(Positron emission tomography) detects the gamma rays emitted by the biologically active molecules and it forms the 2-d and 3-d images of the tracer concentration inside the human body.Glioma titc-SPECT: Spet(Single-photon emission control tomography) or spect imaging requires a gamma-emitting radioisotope (radionuclide) which is injected into the body by the bloodstream. Gamma camera is used to acquire multiple 2-d images from different angles in inSpect imaging technique.Metastatic Adenocarcinoma: It is a type of tumor that spreads from one part of the body to other regions. These cancer cells travel through the bloodstream to other parts of body after getting detached from the main tumor. Once in the blood, they infect other organs of the body. Most of the cells die how so ever few of them settle and again starts to grow as a new tumor in various parts of the body.Metastatic bronchogenic carcinoma: It is a malignant new growth composed of epithelial cells which infiltrate other tissues and give rise to metastases.Sarcoma: It is a Cancerous tumor which arises in the connective tissues. Normal connective tissue includes fat, nerves, blood, cartilage, muscles, deep skin tissues, bones, and vessels. Sarcomas are categorized into two groups, bone sarcomas, and soft tissue sarcomas. They are hard to detect as it can grow in any part of the body Kandwal and Kumar (2014). The symptoms are formation of a painless lump. With the growth of lump, nerves or muscles are pressed and this makes person uncomfortable or give trouble in breathing Bandyopadhyay (2011)-Kharrat et al. (2009).