Read the Conversation
EF: Do you see the year 2023 as a challenge or an opportunity?
ST: 2023 is both a challenge and an opportunity. Four years ago, we started investing in innovation and artificial intelligence for use in healthcare. People are usually skeptical when it comes to artificial intelligence and healthcare. We finished our AI model in 2022 and started offering our services free of charge to public health institutions so that they could use and analyze the system. Once they see it working and accept it, we can get into the real opportunity of officially working with the public healthcare system, with all the certifications and budgets they have planned for us.
It was quite a challenge to get into this market. There was a lot of noise and resistance from other people about existing artificial intelligence models, mammogram research, and whether AI works. We had to prove ourselves, and we did that. We took that challenge, and now there is a huge opportunity in Latin America because there are not enough radiologists, oncologists, and healthcare facilities, and they do not have the advantage of getting this kind of diagnostics easily and cheaply. We have recognized this opportunity, and we are working to cease it.
The change of government and elections next year has created a dynamic atmosphere for companies to establish themselves and sell on a large scale to governments. 2023 is a challenge and a huge opportunity because digital health is a fresh and untapped market.
EF: Could you elaborate on your partnerships? How is your technology assisting with earlier, more accurate diagnostics?
ST: We signed contracts with non-profit organizations and other institutions like university-based hospitals and six other hospitals in Mexico. They are all providing us with roughly 300 patients from each institution. They use our software to identify and diagnose cancer to achieve early detection.
Our system already has the data for 35,000 patients' mammograms and more than 2500 biopsies from cancer patients. The AI model is completely trained. If I feed it data from a mammogram or a biopsy of any of the patients from any population in the world, it has an accuracy rate of more than 95 percent to be able to diagnose cancer.
Another advantage of our system is that it diagnoses a mammogram within one minute, whereas a radiologist takes roughly 10 minutes to evaluate and give his interpretation. We are ten times faster. The system does not get tired, the precision does not get affected, and it is cheaper. We are giving it away for free so that populations that do not have access to that system and insufficient money can take advantage of it. "Life before money."
Our biopsy interpretation happens within 15 seconds. Usually, when a patient goes to a clinic and gets a breast cancer biopsy, it takes more than 15 days to get a result. The patient has to live through hell during that time, whereas our interpretation is within 15 seconds. That gives a tremendous advantage to healthcare professionals who are overloaded with work. They can use this system to get results, prioritize their workload, and give timely responses to their patients.
Our system helps in three ways. Time reduction for attending patients, cost reduction for attending patients, and availability in remote areas because it is a web-based, cloud-based system so that you can use it anywhere in the world.
Beyond that, the advantage is that artificial intelligence trains itself daily with patients from around the world. Its perfection level is improving day by day. The more people use it, the more chances there are of it detecting any kind of anomaly in a fraction of a second. We are projecting that we will have data from the same patients over a longer period in the next three years. We are planning to make a diagnosis before the cancer happens. We do not know the cure, but we know how to detect it before it is too late and help patients and the government reduce the cost of treating cancer.
We are also working with institutions to get genetic data to analyze, based on genetics, the type of cancer, the type of protein, or the type of drug that can help solve the problem or the type of drug that is not working on a certain type of population, thus reducing the cost of treatment for each patient.
We are working in a collaborative model where we want to reduce the cost, increase the chances of survival for the patients, and increase the probability of detection of cancer before it happens.
EF: What has been the public sector's response to adopting AI and technology in healthcare?
ST: Initially, we received a mixed response. When we spoke with one of the largest cancer foundations in Mexico, they were unsure how it would work. A lot of people do not know the precision of AI. People are now looking at it in a very different way. The theories of cancer detection with AI have existed since the 1970s. Still, nobody has invested enough money, effort, or time to gather the data because that is another challenge.
Bringing all those factors together was a huge task. Once the doctors see the results of AI, they become more accepting that this can be an interesting tool that can help them. Earlier, radiologists felt they would not have a role to play and would be replaced by AI. The idea is not to remove the radiologists and replace them with the system. The idea is to give them access to a tool so that they can do their job with much more precision.
When radiologists detect any irregularity in a mammogram, they have to go for a second and sometimes a third opinion. Getting a second opinion from other radiologists can sometimes take a week, and time is of the essence in cancer. Our system can work as a second or third opinion because it has the opinions of more than 35,000 radiologists. It has unparalleled precision. For example, we tested ten Indian patients from a health institute in India. A radiologist had marked none of the data. We were just sent the mammograms, and we developed the results. We were 100 percent accurate with each patient. The radiologists in India were stunned. They had expected us only to diagnose eight of the cases. We accurately diagnosed these cases without having any information about any Indian patients.
The data of Indian patients, such as the type of breast, differs from that of a Brazilian, Portuguese, or Mexican patient. Depending on the region, different types of breast tissues exist within Mexico. AI needs all this very specific data. Our algorithms worked so nicely and came out with this result.
We were so proud and felt so blessed that we could see this day and this result. It gave us so much confidence that we can now go to any population with a precision rate of at least 90 percent.
Healthcare professionals are now more accepting of our system, and we are in talks with the local, state, and federal governments to do a pilot so that they can see the advantages of our system and how they can lower the cost of spending on cancer medicines as well as cancer treatments for the patient. Breast cancer and cervical cancer are the number-one causes of death in women in Mexico, and that is where we are now focused.
EF: Is data security an issue for you? How do you ensure the safety and confidentiality of the data you use?
ST: Data protection is one of the most important factors for the success of any AI-based system. In order to manage data security, we usually give our systems to healthcare institutes, and they run our systems on their cloud and they train our AI with their data.
We do not store their patients’ data on our servers. We just extract the deceased's information but not the patient's information. We have kept these two points separate.
We have also released our electronic medical record system, which will be connected to our AI in the future. This will be useful to those countries that do not have consolidated electronic medical record systems. We can give them everything, help them operate it, and train them to help them solve all types of cancer-related problems.
We also use cloud-based computing to implement this solution, and depending on our customer's choice, we can implement both versions. They can go on Azure or Amazon Cloud, and we can give it to the customer based on their preferences, and their IT teams can operate the whole system.
EF: What are the company's expansion plans?
ST: We do have expansion plans. Our system is open to private sector customers as well as governments. We plan to finish developing our cancer models for other types of cancer. Apart from breast and cervical cancer, we have been working on lung, skin, and prostate cancer. We also want to attempt to model blood cancer and leukemia.
Once we have rolled out all the other models, we intend to work with the central governments of all the countries in Latin America. We are currently in talks with Panama and Peru to see how we can help them implement this model at a national level at a minimum cost so that all these countries, which are lagging technologically in healthcare, can have some advancement.
We also want them to be able to implement the electronic medical record, which is a person's medical identity. With the help of big data and analytics, we want to see how we can work on the root cause analysis of major illnesses with major healthcare costs to help them reduce them. We also want to put our gene sequencers in each country to enable us to do gene sequencing for them using their people and our guide to train them on how to go about it.
We have a vision, and we want our clients to join us in our vision to get the advantage, someday operate on their own, and see how they can help their population with modern technologies and personalized medicine. We want to personalize cancer medicine as much as we possibly can. In the future, we want to go on to diabetes, blood pressure, and all those silent killers to help the population with this technology to reduce the mortality rate.
EF: When you look into the future, where does your vision take you in 10 years? How would you like Brahma to be remembered in a decade?
ST: We will have covered most of Latin America within a decade. Suppose we get enough support in Latin America. In that case, we will be one of the best-equipped AI solutions to help reduce mortality by creating personalized medicine for the entire population.
We want to help reduce the cost of healthcare in all of Latin America so that the budget can be used for other things like education and poverty alleviation. They can use these savings to improve their citizens' quality of life.
We are not a non-profit organization. We must be able to expand and make money. Looking at how the world is going, we will have enough business. So far, we have done it on our own. We have not raised a single penny of capital from anyone. It is all our money so that we will be good in 10 years.
EF: If we were to create a road map for the future of healthcare in Mexico, what would you include as your three main pillars?
ST: We are working with lawmakers in Mexico to train them on the future of healthcare. Technology is coming, and technology has arrived. Technology is beneficial, and we need to capitalize on those benefits.
Education in the public sector is important. We want to educate our lawmakers so they understand the importance of technology, not just AI technology in general. They have to understand and embrace it. Political willingness is the first factor in success.
The second pillar is the budget. There should be money to invest in such technologies. Once you are willing to embrace technology, you must nurture it by investing in it. There must be some sort of road map for investment with concrete goals for what they want to achieve. That must happen within the framework of national health services. They have to embrace it and create a digital framework for how they will implement it. They also need clear objectives that will be used to implement and track the systems.
The third pillar is public awareness. It is important to create awareness about diseases such as cancer by inviting people to go to public health services and get checked. That will give enough data to the National Health Services to analyze and make important decisions on how to change the current situations they are living in.
Political willingness, investment in technology, and educating the public fundamentally change the vision we are planning.
EF: Is there any final message that you would like to mention?
ST: We are looking to start in Latin America and will request cooperation from different governments. We are so willing and happy to work with the central governments in each of these countries to help with our technologies and show them the future of AI in cancer detection and other diseases.
We plan to achieve our goals in the next 5 to 10 years. We want to help these countries and improve the lives of their citizens.
