•  
  •  
 

McNair Scholars Research Journal

Abstract

Health professionals must diagnose correctly to avoid unnecessary costs in care and medicine. Different conditions, come with different treatments. Therefore, it is key to produce the right diagnoses promptly to help treat people appropriately and quickly. As lung cancer is the leading cancer type in the world, it is often diagnosed too late once symptoms appear which becomes difficult to treat at certain stages. In the race against time, recruiting the use of machine learning next to traditional methods of diagnosing will improve the confidence of professionals to predict specific conditions accurately and with speed. For our research, we will develop a convolutional neural network (CNN) model using the TinyVGG architecture to classify subtypes of lung cancer taken anonymously from patients. Pictures will be taken during the 10 days of incubating these cells. Over time, we will look at the results and determine if there are unique overgrowth patterns from different lung cancer cell lines. Alongside this, we will measure the accuracy of the CNN model and its ability to detect features that are not obvious to the human eye. This will aid in the understanding of this model by optimizing and adjusting its parameters to help classify these tumor growths correctly. A simultaneous goal is to determine if these unique overgrowth patterns are correlated with specific activation of certain oncogenes.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.