Improved fuzzy c-means clustering algorithm for automatic detection of lung nodules

Liao, Fan; Zhao, Chenglin

doi:10.1109/cisp.2015.7407925

Cited by 1 publication

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The blue color of solid indicates normal case other than that indicates cancer. Combining both the approaches it is concluded whether the case is normal or abnormal [8].…”

Section: Methodsmentioning

confidence: 99%

“…Each pixel is giving a binary code based on its value relative to its neighboring pixels. This is done by giving a value of 1 neighboring pixel that have values greater than or equal to the value of the pixel being considered and a value of zero to those with lower values [8][9]. This results in an n-bit binary code describing each pixel, where n is the number of neighbors.…”

Section: Dataset Descriptionmentioning

confidence: 99%

“…For the performance analysis of this stage test scans was used. First, we took the results only from the machine learning segmentation approach and followed through the false positive reduction step to calculate the area measures [8]. These three methods provide encodings for the patches.…”

Section: Dataset Descriptionmentioning

confidence: 99%

“…However, we noticed there were still many false positives in the detected nodules, which caused problems when training a classifier. To circumvent this, we develop an approach which involves keeping nodule detections that are found in regions where multiple detections were made across all slices for a single patient [8]. The approach generates a classifier to combine the detections of the Classifier on all the slices for a single patient.…”

Section: Dataset Descriptionmentioning

confidence: 99%

“…If the detected nodule is over a certain value for the dice coefficient, it is kept and cropped to a 50x50 image to keep as a true detected nodule for that patient, and as data for the nodule classification. Moreover, unlike the approach, which averages nodule encodings to arrive at a patient-level representation that corresponds to the provided labels [8], in this approach, a detected nodule is labeled cancerous if the patient has cancer and non-cancerous does not. The advantage of this approach is that, again, it does not rely on detecting nodules.…”

Section: Dataset Descriptionmentioning

confidence: 99%

See 4 more Smart Citations

Lung cancer classification using data mining and supervised learning algorithms on multi-dimensional data set

Ahmed¹,

Al_Barazanchi²,

Mhana³

et al. 2019

PEN

View full text Add to dashboard Cite

These With recent developments in machine learning, data mining and computer vision, there is great potential for improvements in early detection of lung cancer using scans and data available. This paper details the methods and techniques used in our project, where the objective is to develop algorithms to determine whether a patient has or is likely to develop lung cancer using dataset images using data mining and machine learning for the classification and examination. We explore approaches to address the problem. Cancer is the most important cause of death globally. The disease diagnosis is a major process to treat the patients who are affected by cancer disease. The diagnosis process is more difficult comparatively known about the cancer disease detection. Developing a proposed data mining model is useful to diagnose the cancer disease once the cancer detection is accomplished using data mining for the examination and classification of machine learning supervised algorithms.

show abstract

“…The blue color of solid indicates normal case other than that indicates cancer. Combining both the approaches it is concluded whether the case is normal or abnormal [8].…”

Section: Methodsmentioning

confidence: 99%

Section: Dataset Descriptionmentioning

confidence: 99%

Section: Dataset Descriptionmentioning

confidence: 99%

Section: Dataset Descriptionmentioning

confidence: 99%

Section: Dataset Descriptionmentioning

confidence: 99%

See 3 more Smart Citations