Prostate Analysis Automation

Welcome 

This is a project where we explore ways of improving tissue sample extraction on behalf of MR-Labs at ST. Olavs Our goal is to make a solution that is easy to use, effective and provides high quality samples.

The project can be divided into several sub-problems:

The Cutting/Drilling

Current procedure:

Manually lowered stationary hole-drill with manual moving plate. Drill-bits are created specifically at NTNU for this purpose.

Possible improvements:

• Automatic cutting
• Variety of sample shapes and sizes.

Obstacles:

• Vibration
• Temperature
• Brittleness

Tissue Recognition/Mapping

Current procedure:

Manual Recognition by medical personnel by using MR photos and photos of each side of the slice.

Possible improvements:

• Computer Vision?
• Difference in density (Cancer cells is more dense)
• Hyperspectral or Terrahertz-camera

Obstacles:

• Impossible to see cancerous tissue on frozen prostate with human eye (even for trained doctors)

The Gripping/Clamping

Current procedure:

Prostate slice is pushed from four sides by clamps. This is problematic as the clamps sometimes slip and the operator has to readjust them.

Possible improvements:

• Smaller clamping area
• Thousand-needle concept
• One-time-use concept

Obstacles:

The Planning

Current procedure:

Usually takes several hours.

Possible improvements:

Obstacles:

• Not very high priority

Temperature exposure

Current procedure:

Possible improvements:

• Regulate temperature precisely (Temperature-fluctuation is ok as long as it happens below limit.)

Obstacles:

• Humidity from the frozen tissue
• -10 is absolute limit, defrosting destroys sample

Minimize human interference

Current procedure:

• Everything is done by human, From slicing to drilling

Possible improvements:

• Automate the procedure, so that all the human has to do is place the slice in a drawer, and the finished test samples and the prostate slice comes out in seperate drawers, ready for storage.
• Make the technical aspect of the procedure easier for the operator.

Obstacles:

• A lot of smaller procedures must be automated before the whole procedure can be automated.

Contamination

Current procedure:

Possible improvements:

Obstacles:

• Avoid polymer contamination

Time reduction

Current procedure:

• Planning the drilling: Hours to days
• Actuall Drilling: 10-15 minutes (20s per sample, 10-30 prøver per slice)

Possible improvements:

• Automation of drilling procedure
• Computer vision

Obstacles:

Temporary Storage of Samples/Transportation

Current procedure:

Manual moving of cut sample from cutting board to specimen tube.

Larger sample glass. 

Possible improvements:

• Shute/Fall into tube below

Obstacles:

• Has to stay cooled