Go to lewagon.com

Exploring the Potential of AI in Healthcare with Daniela Rodrigues

In honour of World Health Day, we spoke with Daniela Rodrigues, a Le Wagon alumni. With a desire to combine her medical experience with emerging technologies, Daniela joined our Data Science bootcamp. Read her interview to learn more about her interesting insights on the potential of AI in healthcare!

“I had heard about the potential of AI in healthcare, especially in the field of medical imaging, and it sounded super exciting to me. However, I lacked the expertise to perform such a job.”

Hi Daniela, can you please introduce yourself?

Hi, my name is Daniela. I hold a Master’s degree in Medicine and worked as a Pulmonology Resident for four years. However, last year I decided to change my career path and move into the field of Data Science and Artificial Intelligence. My ultimate goal is to apply my medical knowledge to this exciting new field.

Before Le Wagon, you have been practicing medicine. What made you decide to pursue a Data Science bootcamp?

I had been considering a career change for some time, as my current job wasn’t very fulfilling. Most of my time was spent on paperwork and dealing with logistic issues, which is common among doctors nowadays, instead of actually seeing patients. I had heard about the potential of AI in healthcare, especially in the field of medical imaging, and it sounded super exciting to me. However, I lacked the expertise to perform such a job. After conducting extensive research, I realized that attending a bootcamp could provide me with the basic skills necessary to start working on projects and begin my new career.

Could you tell us more about your post-bootcamp experience?

After completing the bootcamp, I was fortunate enough to promptly secure an internship at a medical imaging company, which aligned with my field of interest. The internship was part of a project called innovAId, initiated by the European Institute of Innovation and Technology to promote innovation in digital health. My work focused on deep learning for computer vision and image processing techniques.

Can you describe your experience with implementing AI in healthcare?

As previously stated, I have only worked in the field of medical imaging. During my internship, I developed a project for classifying lung cancer using deep learning, which drew on my previous experience working in pulmonology. Currently, I am still employed by the company, but I am focused on developing their main product.


How do you see AI improving patient outcomes in medicine?

There are numerous ways in which AI could improve patient outcomes while also making doctors’ lives easier. For instance, in the detection of lung cancer, an algorithm that could predict whether a lung nodule would develop into cancer could allow for early treatment and significantly improve survival rates. Additionally, this approach would help avoid the need for several follow-up CT scans and consultations, thereby reducing patients’ exposure to radiation and conserving hospital resources. This is just one example, but the same approach could be applied to many other areas of medicine.

What are some of the ethical considerations that come with using AI in medicine, and how do you address them?

Patient privacy is a crucial consideration when using AI. Large amounts of data are necessary for accurate predictions, but this may also put patient privacy at risk if the information is used for a different purpose. Therefore, it is important to have strict policies for data protection. Patients should also be informed about how their information will be used and give their consent.
Another ethical issue regarding AI is its use in clinical practice. In my opinion, AI should only be used as an additional tool to help doctors with their decisions and diagnoses, and never as the sole decision maker. It could be included in guidelines, but the final decision should always be made by the doctor.

What challenges have you encountered in implementing AI in medicine, and how have you overcome them?

One of the biggest obstacles in machine learning is obtaining sufficient data to train the models. Patient privacy is paramount, making access to information more difficult. Additionally, obtaining data often requires expertise from skilled professionals. For example, in medical imaging, annotations made by doctors are often necessary, which is both expensive and time-consuming. As a result, most datasets available in medicine are not very large, which is not ideal for training algorithms. Fortunately, there are techniques that can be used to overcome this issue, and I have applied them in my project.

How do you see the field of AI in medicine evolving in the next 5-10 years?

AI is evolving at an incredibly fast pace. It has the potential to improve patient care, make doctors’ work easier, and improve resource management. However, there is no doubt that AI should be highly regulated to avoid its misuse and potential harms.

Thank you Daniela!

Our users have also consulted:
Pour développe mes compétences
Formation développeur web
Formation data scientist
Formation data analyst
Les internautes ont également consulté :

Suscribe to our newsletter

Receive a monthly newsletter with personalized tech tips.