Skip to content

R3plica

AI In The Battle Against Cancer

AI, Machine Learning, Health Care3 min read

Maheshwar Kuchana

Meet Maheshwar Kuchana

AI Engineer @ Adventum

Bengaluru, India 

Maheshwar Kuchana works on AI for Healthcare to make an impact in the rural Indian societies.

He had done a couple of researches on Cancers, COVID-19 Diagnosis, Melanoma, IVF success rate, and more. He enjoys creating things that live in the real world, whether that be AI models, applications, or anything in between.

His goal is to always build products that provide robust, practical, real-time experiences.


What inspired you to pursue a career in Artificial Intelligence?

Applications of AI starting from face detection to diagnosing a disease have deeply impacted the curiosity in me and made me crave more knowledge in the field of AI.

The first project I made in AI was on Fingerprint Recognition using Machine Learning and it came out to research in employing a new technique to find Region of Interest.

Having experienced in that project, opened a gate for interest in AI inside me. From 18 (age), I am fascinated by Biology and after learning a few bits of AI I wondered if there was any application of AI on healthcare.

To my surprise, there was a lot of work done on healthcare on AI.

Knowing that people have made a boilerplate for me I started doing my research on AI for Healthcare and started building applications in it.

How Deep Learning could help in the Cancer diagnosis?

Cancer is the leading cause of death worldwide. Both researchers and doctors are facing the challenges of fighting cancer.

The diagnosis of cancer depends on what tests must be performed for a type of cancer. Essentially, it means for some cancers X-rays, CT, MRI, biopsy tests must perform, for some only a few tests need to be performed.

The main challenge where AI must be placed in finding radiological features, concluding it as particular cancer.

Machine learning helps pathologists as they try to interpret biopsy slides and determine whether a patient’s breast cancer has metastasized to nearby lymph nodes.

While deep learning eases the process of finding features doing it automatically. Recent deep learning architectures help in diagnosing cancers very efficiently that perform on par with the human level.

How AI could detect early stages of Lymphoma cancer?

One of the ways where AI could detect early stages of Lymphoma Cancer is by analyzing Histopathological images (microscopic examination of tissue).

Histopathological diagnosis of lymphomas represents a challenge requiring either expertise or centralized review and greatly depends on the technical process of tissue sections.

Whole-slide images of lymph nodes affected by FL or follicular hyperplasia will be used for training, validating, and finally testing Deep neural networks (DNN).

Lymphoma diagnosis depends on the expertise of the pathologist who, in the case of follicular proliferation, has to clearly distinguish follicular lymphoma (FL) from follicular hyperplasia (FH), both being lesions that sometimes display very similar features.

Deep learning-based diagnostic systems have recently provided automated methods for histopathology image analysis, which may reduce inter- and intra-observer variability in cancer diagnosis through an objective analysis.

In simpler words, it helps us to classify using complex Convolutional Neural Network architecture whether as Cancerous or Non-Cancerous tissue.

Do you have any Ai project that wants to share your experience with us?

In recent days, I along with three other AI enthusiasts have developed a production-ready, open-sourced diagnostic tool for COVID-19.

As the diagnosis of COVID-19 became challenging with RT-PCR and Antibody tests the scientific/medical community has suggested including radiographs (CT, X-ray) in diagnosis. And very soon after the outbreak, there are 2-3 sources where the X-ray, CT scan data along with clinical information has been open-sourced.

We have developed a diagnostic that can detect abnormal conditions in lung CT scans and X-rays using Deep Learning. Not just diagnosis but our tool provides a facility to track patient recovering progression i.e., tracking patients. We have been funded by an organization by looking at version 1.

Currently, the solution is used in a hospital that is in Mumbai, India in diagnosing patients.

Where AI could have a big impact? and Why?

The Healthcare industry could have a big impact on AI. Artificial intelligence has the potential to transform how healthcare is delivered.

AI can be used to explore how it can support improvements in care outcomes, patient experience, and access to healthcare services. It can increase productivity and the efficiency of care delivery and allow healthcare systems to provide more and better care to more people.

AI can help improve the experience of healthcare practitioners, enabling them to spend more time in direct patient care and reducing burnout.

We are in the very early days of our understanding of AI and its full potential in healthcare, in particular with regards to the impact of AI on personalization.

But it’s a broad application that covers natural language processing (NLP), image analysis, and predictive analytics based on machine learning.


For more about Maheshwar Kuchana projects go to maheshwark.com