Detection of Cancer in its Early Stages by using CNNs

[varchanaiyer]

Abstract

There is a strong correlation between a person's DNA and the diseases they might have. This becomes especially important in the case of Cancer where traditional methods to detect cancer can take days and may not be accurate. In this talk, I will show you how you can train a CNN to detect cancer using genetic data. Since DNA data can be huge, this talk will also be about how you can optimize your code to handle large datasets (>200 GB) without it being a bottleneck during your training.

About Part 1 - What DNA data looks like. How DNA is related to cancer. How to obtain open-source DNA data Part 2 - How you go about starting an AI project: Obtaining data, Preprocessing data, training a model, improving the model, testing and inference Part 3 - How to make Data Preprocessing faster using numpy, dask and numba. How to deal with large datasets and how to store it in memory. Part 4 - Demo of the techniques Part 5 - Implementing papers and creating your own models using TensorFlow Part 6 - Demo of the model and implementation Part 7 - Concluding with future possible work and resources to get started. Part 8 - Q/A

Pre-requisites

-Elementary knowledge of Deep Learning -Python and TensorFlow

Expected duration

The time required for the talk: The talk will be of 30 mins with 5 mins of Q/A in the end. The following are the parts to it: DNA Data and Cancer Introduction - 1 min The process of working on an ML project - 3 mins Data, Data Preprocessing and Dealing with large datasets - 10 mins Model Architecture and Demo - 10 mins Conclusion - 1 min Q/A - 5 mins

Level

Intermediate

Resources

-Slides can be found at here -The paper published can be found here

Speaker Bio I am a fresher from SRM Institute of Science and Technology. I understand that engineering is not everyone’s cup of tea and that everyone has a different perception of it. During my second year of study, I realized that for me education was something that was present beyond books and into practical applications. So I collaborated with a few other mates in college and started this place called the Next Tech Lab which was involved in cutting-edge innovation and novel research ideas.

As a few of my achievements that the lab made me achieve included winning the Smart India Hackathon 2017 as the first prize under Ministry of Steel for using machine learning to detect power theft in India. Recently I was invited to the WiPDA conference in Xi’an China for presenting my work in GaN modeling of devices using machine learning, a collaboration with the University of Cambridge. I have around 3 IEEE Xplore Papers and 1 Elsevier papers for my contribution to electrical and machine learning fields

As a lab, we have done so much more to protect gender diversity even among the strength of 200 members keeping a ratio of 50:50. We were portrayed for accomplishments by the News 18 in a short video. Over the past 6 months, I have had the opportunity to work and intern at Saama Technologies where I research on Machine Learning in order to accelerate clinical trials. A part of this work has exposed me to how machine learning models are necessary to be used in various genomics fields.

You can view my Google Scholar citations here

You can view my blogs

1) Women and Data Science 2) Quantization and need for TPUs 3) Application of signal processing in machine learning

[im-gozmit]

Hi @ArchanaIyer1996, Thanks for proposing the talk. Are you available for delivering the talk on the 23rd of September?

[varchanaiyer]

Hi! I don't think I will be available this weekend since I work from Chennai. But I do come to Delhi on a regular basis. Maybe you can send me your schedule for the next 2 months? That way I can travel my plan accordingly.

[im-gozmit]

Okay, so @utkarsh2102 can you please provide her the Schedule for the upcoming meetups.

[utkarsh2102]

Sure, will ping accordingly.

[varchanaiyer]

Hi, @utkarsh2102 I will be in Delhi from the 27th October to 13th November! Let me know if you have anything coming up then, will for sure make it. Thanks!