Cancer Survival Calculator

Cancer Survival Calculator 

Dr. Lauren Janczewski at Northwestern University McGaw Medical Center in Chicago, alongside a team of researchers, has developed an AI-based tool named the Cancer Survival Calculator to estimate the survival chances of newly diagnosed cancer patients. The study aimed to identify the patient, tumor, and treatment characteristics that greatly influence patient survival for each cancer type.

The AI was tested on a large dataset of 420,623 patients from the National Cancer Database (259,485 breast cancer, 76,624 thyroid cancer, and 84,514 pancreatic cancer). The tool has shown remarkable accuracy in forecasting five-year survival rates for cancer patients. 

This is a leap forward in personalized cancer care.

News coverage: News Medical
Doctors: Dr Lauren Janczewski
Institutions mentioned: 
Northwestern University McGaw Medical Center, Chicago

The AI Platform C the Signs Predicts Cancer With Up To 97.5% Accuracy

The researchers tested the AI on 111,421 patients in the UK, and the tool identified 7257 patients at high risk. In reality, 7360 patients were diagnosed with cancer. 

The AI platform was most accurate for gastrointestinal tumors (97.5% accuracy), skin cancer (96.9%), breast cancer (96.6%), gynecologic cancers (95.3%), urologic cancers (94.0%), brain and central nervous system cancers (92.3%), upper gastrointestinal cancers (87.0%), and chest tumors (86.8%).

The AI also correctly identified the tumor origin in 93.2% of the cancer patients.

This type of AI cancer prediction platform will help with earlier cancer detection and to accurately detect tumor origin in patients with vague symptoms and facilitate targeted investigations appropriate to the tumor type. 

Link to research paper: American Society of Clinical Oncology
News coverage: Cancer Therapy Advisor
Doctor’s mentioned: Dr Bea Bakshi
Institutions mentioned: C the Signs (Company)

An AI-driven Digital Pathology Colorectal Cancer Diagnostic Tool

Colorectal cancer is the third most common cancer globally and the second highest cause of cancer mortality. French-American company Owkin has developed MSIntuit™ CRC, an innovative artificial intelligence tool designed to improve the accuracy of detecting and treating colorectal cancer.

Microsatellite Instability, or MSI for short, is a sign doctors look for in colorectal cancer, which affects about 15% of patients. Knowing if a patient's cancer has MSI is important because it helps predict how the cancer will progress and how it might respond to certain treatments. Recently, a type of treatment called immune checkpoint inhibitors has been shown to work well for patients with MSI. So, doctors are now using MSI testing as a standard part of deciding the best treatment for patients with colorectal cancer.

This AI powered tool can pinpoint nearly half of the patients who don’t have a specific cancer feature, while accurately identifying those who do with over 96% accuracy. This breakthrough could help doctors make faster, more informed decisions and get effective treatments to patients more quickly. It also promises to cut costs and streamline testing procedures, especially in places where testing is done on a large scale.

Link to research paper: Nature Communications
News coverage: News Medical
Doctor’s mentioned: Magali Svrcek
Institutions mentioned: 
Saint Antoine Hospital
Owkin, Inc. (Company)
Medipath

The UK allocated £21 million ($26 million) to roll out AI tools for lung cancer  

In the UK, £21 million is being allocated to 64 NHS trusts across England to roll out AI tools to speed up the diagnosis and treatment of lung cancer. 

The AI tools are used for analyzing X-rays and CT scans. With over 600,000 chest X-rays performed each month in England, the investment will help doctors with quicker and more accurate diagnoses of conditions. 

The UK is taking great strides in implementing AI in healthcare, which we have already seen with Brainomix, which is helping doctors halve treatment times for stroke patients. I will talk about Brainomix next week, but you can read my coverage of Brainomix in my Practical AI newsletter here. 

Link to press release: UK Government
Doctor’s mentioned: -
Institutions mentioned: 
NHS Trusts
Brainomix