AI model identifies certain breast tumor stages likely to progress to invasive cancer.
A new machine-learning model developed by researchers from MIT and ETH Zurich can identify the stage of disease in ductal carcinoma in situ (DCIS), a type of preinvasive tumor that sometimes progresses to a highly deadly form of breast cancer. DCIS accounts for about 25 percent of all breast cancer diagnoses and is often overtreated due to the difficulty in determining its type and stage.
The AI model uses cheap and easy-to-obtain breast tissue images to identify the different stages of DCIS. The researchers built one of the largest datasets of its kind, containing 560 tissue sample images from 122 patients at three different stages of disease, to train and test their model. The model learns a representation of the state of each cell in a tissue sample image and infers the stage of a patient’s cancer based on this information.
The researchers found that both the state and arrangement of cells in a tissue sample are important for determining the stage of DCIS. They designed the model to create clusters of cells in similar states, identifying eight states that are important markers of DCIS. Some cell states are more indicative of invasive cancer than others. The model determines the proportion of cells in each state in a tissue sample and considers the spatial organization of these cells, which significantly boosted its accuracy.
When compared to samples evaluated by a pathologist, the model showed clear agreement in many instances. In cases that were not as clear-cut, the model could provide information about features in a tissue sample, like the organization of cells, that a pathologist could use in decision-making. This versatile model could help clinicians streamline the diagnosis of simpler cases without the need for labor-intensive tests, giving them more time to evaluate cases where it is less clear if DCIS will become invasive.
The researchers believe that this model could be used as a tool to help clinicians assess breast cancer stage and ultimately help in reducing overtreatment. They plan to conduct a prospective study, working with a hospital to get the model all the way to the clinic, which will be an important step forward.
