Person Doing Blood Test

SolVisionCase Study

Pharmaceuticals and Medical

AI Visual Identification and Classification of Cells

AI enabled cancer cell detection 

Cell inspection in pathological examinations

During cancer screening, specimens are obtained through biopsy or liquid biopsy, and can contain cells shed from the tumor or DNA related to the tumor. Doctors determine whether cells are normal or unhealthy based on their unique characteristics, and refer to the degree of cytopathy to determine prognosis and subsequent treatment. 

AI enabled cancer cell detection

Biopsy images and cell variations

Analyzing cell characteristics can distinguish cancer cells from normal cells and also classify cancer cells. However, under microscopic images cancer cells vary in appearance and are randomly located, which could drastically affect a doctor’s judgment or selection criteria. In light of these uncertainties, traditional inspection systems are insufficient for detecting and determining cell variations. 

Classifying cells with AI deep learning

Using SolVision’s Classification tool, the AI model can be trained to recognize the different features of normal cells and cancer cells. The system’s data augmentation tool enables simulation of potential cell types and variations to strengthen training and stability of the AI model. The advanced AI model can then be used to identify and classify cells accurately for doctors to make better prognosis. 

AI Inspection

Normal cells

Cell lesion imaging test case

Mid-stage lesion

More Case

  • AI Defect Detection for Spiral-Surfaced Metals

    Small, spiral-surfaced metal parts can be inspected using SolVision’s Instance Segmentation tool to learn the different types of cut marks or collision faults from sample images, then building an AI model to recognize these subtle defects.
  • male professional checking the condition of a fire extinguisher

    Fire Extinguisher Inspection Using AR + AI

    Discover how a company improved fire extinguisher inspections by implementing META-aivi, enabling real-time issue detection, streamlining processes, and reducing inspection time by 60%.

  • AI Inspection for Product Labels

    Using AI deep learning, Solomon SolVision’s Instance Segmentation tool can learn the many types of text and number defects on labels, even if they are irregular or subtle.

  • AI Quality Control for Medical Masks Manufacturing

    SolVision’s deep learning inspection software reduces the number of defects to almost zero during medical mask production