Imagine if you were just diagnosed with bone cancer, one of the most dangerous forms of the most unpredictable disease. The doctors put you under and cut a piece of your bone to study its genetic code. “But how could they use this to get rid of my cancer?” you may ask, “How will they turn this into a magical antidote?” Well, it could be used in combination with Artificial Intelligence to generate a diagnosis and treatment for you.
Many people think of AI as something that they use to get ideas for an essay, get answers for a math problem, create vacation ideas, or find out what bug is on the ground. While this may be true, it does not mean that is AI’s only purpose. AI is now being deployed to hospitals around the world for imaging, pathology, drug discovery, genomics (the study of an organism’s DNA and genes), and more. The AI models actively use information from you and others alike to find patterns that are used to detect cancer before it becomes dangerous, and speed up research done on medical conditions.
The basics of artificial intelligence are easy to understand. AI uses algorithms to learn patterns from data that it is given to make predictions or assist a user’s decision making (It uses outside info to suggest ideas). In the medical field, it uses data from CT, MRI, mammogram scans, pathology slides, genomes (an entire set of DNA information), clinical notes, and treatment records. It’s being used to identify patterns that are too subtle, complex, or time-consuming for humans to process by themselves.
AI models compare video and data evidence to the data they have collected and use it to detect cancer faster and more accurately than humans ever could. It also uses this data to create treatment plans for patients that are safe and effective. Cancer care generates a massive amount of data from each person, and using this data is a natural fit for AI. Check-ins and monitoring can also be given to AI models to detect cancer relapses and get rid of them as fast as possible.
AI adoption is accelerating greatly, and for good reasons. It is easier than ever before for these AI models to use data from patients because it is all digital. AI can be easily trained because the information presented is ready to be accessed. Incredibly fast computer parts are being made every year, making access to fast and accurate AI tools easier than ever before. Rising cancer rates are also contributing to the increase in AI adoption. Because AI has been proven to be very useful in the medical field, more hospitals are using them.
AI is also improving radiology workflows, making information faster to read, and increasing the sensitivity of breast cancer screening. Information from a study by Moffitt Cancer Center reveals that AI tools are reducing false positives and making turnarounds faster. A quote from The Lancet, a digital health journal, says, “As of July, 2025, over 1200 AI-enabled medical devices have received clearance from the US Food and Drug Administration (FDA). However, only two, Paige Prostate and Galen Second Read, have been authorized for use in histopathology.”
“Our research shows that AI can be a powerful tool for doctors,” said Dipesh Niraula, Ph.D., an applied research scientist in Moffitt’s Machine Learning Department. “But it’s important to recognize that AI works best when it’s used as a support, not a replacement, for human expertise. Doctors bring their expertise and experience to the table, while AI provides data-driven insights. Together, they can make better treatment plans, but it requires trust and clear communication.”
AI systems examine digital pathology slides, which allows them to spot cancer much faster than manual review, which is the traditional way hospitals identify cancer. According to The Lancet, “…these models can also detect morphological signatures predictive of molecular alterations and clinical outcomes,” making pathology both quicker and more accurate.
In addition, AI can analyze genomic data by using scans and identifying paths that guide personalized cancer treatments. ScienceDaily highlights studies showing that these models can analyze complex patient data to help doctors decide which therapies are most likely to work. This can increase the speed of treatment progression and help recovery after the treatment while helping these models learn at the same time.
Evidence from Moffitt Cancer Center derived from a study orchestrated by Nature Communications demonstrates that AI tools analyze complex clinical and imaging data in real time, leading to measurable improvements like faster diagnostic turnaround, more accurate tumor characterization, and better patient stratification for treatment. Data reported by the AACR also shows that this approach can identify patients most likely to benefit from targeted therapies.
Black-box AI models create challenges because they have less transparency in what decisions are created from, making it hard for doctors to understand why these decisions are made. These opaque processes can cause uncertainty and reduce trust, especially when the models behave unpredictably in unfamiliar/rare cases. Information cited from CancerResearch stresses the importance of transparency and human oversight, so clinicians can validate these recommendations with their own professional judgment while keeping them safe and reliable.
With rising workforce shortages in radiology and pathology, hospitals are turning to AI tools to automate repetitive diagnostic tasks and lower the delay that happens for care of patients. Reports show that AI can help fill these gaps by pre-screening scans, flagging urgent cases, and prioritizing helping physicians that have heavy workloads.
Experts from the PMC review and recent reporting from Reuters show that the rapid expansion of medical AI needs to have regulations and guidelines to keep patients safe. They highlight the importance of addressing unfair algorithms, data governance, model reliability, and more. These steps are beneficial for ensuring that AI keeps helping patients the correct way, and not worsening conditions and introducing new risks.