Deep learning

DL Takes on Tumours

In brief:  A cell biologist at Åbo Akademi University in Turku, Finland, Guillaume Jacquemet,  is using DL to track the nucleus of each cell in frame after frame of time-lapse microscopy films. He has trained a machine to spot the nuclei using methods available on a platform called ZeroCostDL4Mic, part of a growing collection of resources aimed at making AI technology accessible to bench scientists who have minimal coding experience. A major driving force has been the explosive growth of life-sciences data. With modern gene-sequencing technologies, a single experiment can produce gigabytes of information. The Cancer Genome Atlas, launched in 2006, has collected information on tens of thousands of samples spanning 33 cancer types; the data exceed 2.5 petabytes, and advances in tissue labelling and automated microscopy are generating complex imaging data faster than researchers can possibly mine them. 

Why this is important: Jacquemet said himself: “If I had to do the tracking manually, it would be impossible.” As we have previously seen, researchers are training a DL model to predict drug responses on the basis of a person’s cancer-genome sequence.  AI is having a profound effect on the future of cancer detection and treatment.