LUNG CANCER DETECTION USING KNN ALGORITHM

• 

  Performance matrix analysis

• 

  performance score

Inspiration We got inspired by the thoughts of our teacher on lung cancer and the urge to prevent it.

What it does

This paper describes the development of a lung cancer detection system using image processing techniques. The created system can capture all medical images consisting of CT images. The model proposed here is developed using DCT for trait selection and SVM, Random forest, KNN, linear regression, logistic regression and Naive Bayes classifiers used for classification. This paper is an extension of image processing with lung cancer detection and presents the results of feature extraction and feature selection after segmentation. The system accepts as input any medical image of CT. In this research, we propose a method to efficiently detect cancer cells from CT images. Super pixel segmentation is used for segmentation and Gabor filter is used for denoising medical images. In the cancer detection system, We use GOOGLE COLAB that helps us in obtaining simulation results for each classifier comparing medical images providing us , Accuracy , Precision , F1 Score, MCC and Error Rate.

How we built it we built in by analyzing various test and train images and implementing various optimization techniques in order to optimize the images and improve our accuracy. we did our project entirely on GOOGLE COLAB.

Challenges we ran into we faced a lot of challenges in both obtaining the training and testing images and also in identifying the best algorithm among various algorithm which provides high accuracy rate and less error rate

Accomplishments that we're proud of we found than KNN algorithm provides about an accuracy of 99.56% which is a great achievement. from now on implementing our method will detect lung cancer more accurately and save more lives.

What we learned we learned a lot of concepts in python regarding machine learning and also came to know about lung cancer deeply which showed us a major part of a sad life and we intend to learn and improve our system and also find new methods to provide solution for cancer

What's next for LUNG CANCER DETECTION USING KNN ALGORITHM In the future scope, the accuracy of the system can be improved if training is performed by using a very large image database. Similarly, we can use CT scan images to detect different types of lung cancer or other cancers using the same approach. We tent to find and improve the accuracy of the results with other classifiers.

Built With

• google-colab
• knn
• machine-learning
• python
• svm