Home » Blog » AI Bio-Imaging Bank for Cancer Detection

AI Bio-Imaging Bank for Cancer Detection

By Amigos IAS

Why is it in the news?

  • Tata Memorial Hospital (TMH) in Mumbai, India’s largest cancer hospital, has introduced AI with a ‘Bio-Imaging Bank’ for early-stage cancer detection.

Bio-Imaging Bank

  • Contains radiology and pathology images along with clinical details, outcome data, treatment specifics, and additional metadata.
  • Purposefully designed for training, validation, and rigorous testing of AI algorithms.
  • Utilizes deep learning to create a cancer-specific tailored algorithm for early detection.
  • It involves data from 60,000 patients.
  • Primarily addresses head and neck cancers and lung cancers; Aims for a minimum of 1000 patients for each cancer type.
  • Targets tasks like screening for lymph node metastases, nucleus segmentation, classification, and predicting biomarkers (e.g., HPV in oropharyngeal, EGFR in lung cancer).
  • Funded collaboratively by the Department of Biotechnology, IIT-Bombay, RGCIRC-New Delhi, AIIMS-New Delhi, and PGIMER-Chandigarh.

AI for Early Cancer Detection

  • Analyses radiological and pathological images, learning from extensive datasets to identify unique features associated with different cancers.
  • Enables early detection of tissue changes and potential malignancies.
  • TMH teams segment and annotate images, correlating them with biopsy results, histopathology reports, and genomic sequences to develop algorithms.

Significance

  • AI-driven early diagnosis has demonstrated a 98 percent correctness rate after cross-verification by doctors.
  • AI’s potential lies in tailoring treatment approaches based on diverse patient profiles, thus optimizing therapy outcomes.
  • Reduces radiation exposure for paediatric patients undergoing CT scans by 40%.

Signup for newsletter

Receive notifications straight into your inbox

Leave a comment

Item added to cart.
0 items - 0.00

Discover more from AMIGOS IAS

Subscribe now to keep reading and get access to the full archive.

Continue reading