A team of Google researchers has used a deep-learning algorithm to accurately predict lung cancer with a calculated scan.
This work demonstrates the ability to enhance both the accuracy and stability of Artificial Intelligence (AI), which can help accelerate the investigation of lung cancer worldwide.
According to the World Health Organization, lung cancer is the most fatal of all cancer in the world – more than breast cancer, prostate and colorectal cancer – and it is the sixth most common cause of death worldwide.
“Shravan Shetty, M. S.”, using the progress in 3D volumetric modeling with datasets as well as datasets from our partners (including Northwestern University), we have progressed in the prediction of modeling lung cancer. Technical Lead told Google in a blog post late last night.
Google researchers created a model that can not only produce malignant prediction (seen in 3D quantities) of overall lung cancer, but also can detect micro-fatal tissues in the lungs (lungs body).
In research, Google AI took advantage of 45,856 de-identified chest CT screening cases (some found in cancer).
“When a single CT scan is used for diagnosis, then our model has performed better than or equal to six radiologists. We compared 11 percent of cancer in the wrong-positive exams compared to unwanted radiologists in our study. Detected cases. “Google said.
For a heterogeneous patient with the history of cancer, the AI system reviewed and investigated potential lung cancer, which was previously called normal.
These initial results are encouraging, but further studies will assess the impact and utility in clinical practice, Google said.