AI is growing into oncology labs
AI tools are moving from proof‑of‑concept into cancer research and diagnostics, but presenters and researchers are stressing that machines won’t replace humans. Market forecasts point to strong AI growth in life sciences, while companies will be showing AI‑driven spatial transcriptomics and pathology tools at AACR alongside posters arguing that clinical judgement still matters. That mix means jobs will increasingly sit at the intersection of software, biology and clinical practice — engineers, computational biologists and clinicians working together. (biospace.com, citizentribune.com, indicalab.com, newswise.com)
Cancer labs still start with glass slides and tissue stains, but more of the reading is now being handed to software that can count cells, trace shapes, and flag patterns too small for a human eye to tally by hand. At the American Association for Cancer Research annual meeting in San Diego on April 17-22, 2026, companies and research groups are showing that shift in real time. (aacr.org) A pathology slide is basically a microscope photo of a tumor sample. Digital pathology turns that photo into a giant file that software can search the way map apps search a city block, looking for every nucleus, membrane, and immune cell in the image. (indicalab.com) Spatial transcriptomics does something similar for gene activity. Instead of just asking which genes are on inside a tumor, it maps where those signals sit, so researchers can see which cells are talking to each other and which ones are hiding from the immune system. (prnewswire.com) That is where artificial intelligence fits in. These systems are being trained to sort huge image files and gene maps fast enough to find patterns linked to drug response, tumor neighborhoods, and possible biomarkers for new medicines. (prnewswire.com) One sign this is moving past the demo stage is the trade-show floor itself. Indica Labs says its HALO and HALO AI software will be featured in multiple American Association for Cancer Research 2026 posters, including studies on stem cell markers, renal cell carcinoma tissue, and multiplex imaging of breast and gastrointestinal cancers. (indicalab.com, aacrjournals.org) Another sign is the scale of the claims. Portrai said on April 8 that it will present 11 posters at the meeting, covering artificial intelligence agents, foundation models, and spatial biomarkers aimed at decoding the tumor microenvironment for oncology drug discovery. (prnewswire.com) Money is following the same direction. A market forecast circulated through BioSpace this week said artificial intelligence in life sciences could grow from $3.27 billion in 2026 to $15.94 billion by 2035, which would be a 19.30 percent compound annual growth rate. (biospace.com) But the people presenting this work are not describing a robot pathologist that replaces doctors. Yale Cancer Center said its teams will bring research to the same meeting that includes a poster arguing that artificial intelligence support in breast pathology still leaves room for clinical judgment by physicians. (medicine.yale.edu) That balance is the real story inside oncology labs right now. The machine is getting good at scanning millions of pixels and thousands of gene signals, while the human is still the one deciding whether a weird pattern is a treatment clue, a technical artifact, or a dangerous false alarm. (medicine.yale.edu, indicalab.com) So the jobs growing around cancer research are not just “biologist” or “software engineer” anymore. They are computational biologists who can work with gene maps, pathologists who can validate algorithm outputs, and engineers who can turn lab images into tools doctors will actually trust. (aacr.org, biospace.com)
