Dr. Greg Clement and Dr. Kullervo Hynynen of Harvard Medical School have devised what they believe is the world's first clinically feasible way of carrying out non-invasive ultrasound surgery across the skull. As well as removing tumors, the technique could possibly be used to deliver drugs to targeted areas in the brain.

Sending ultrasound into the brain is difficult because as the sound wave passes though the skull it scatters (bounces off) and attenuates (gets weakened), making it hard to focus. In the last few years there has been much research to try to assess whether it's possible to use focused ultrasound to carry out brain surgery and therapy.

Work has concentrated on three aspects:

* Demonstrating that ultrasound can, in fact, be sent through the skull and be reliably focused;

*Investigating the skull heating and heating of other areas away from the focus that this causes; and

* Trying to achieve a focus intense enough to coagulate tissue safely in a specified region.

Recent work by the authors and their colleagues has shown both theoretically and experimentally, using human skulls donated for medical research, that focusing through the skull in a precise location without excessive heating is possible.

This can be done by distributing the ultrasound over the entire upper surface of the skull by using a helmet-like hemispherical array of transducers. The final step necessary was to devise a reliable, minimally invasive -- or, ideally, completely non-invasive -- method of treatment.

Now, Clement and Hynynen have taken that step. Their new, non-invasive focusing method uses CT (computed tomography) images to predict how the ultrasound will move through the skull. Understanding how the sound waves will be reflected, diffracted and absorbed by the skull requires information about the skull's thickness and internal structure, as well as precise registration between all points of the skull and the ultrasound array.

Working out exactly how to deliver the required ultrasound also needs to be kept as computationally uncomplicated as possible -- but without oversimplifying the problem.

The algorithm developed by the researchers uses thickness, density and directional information obtained from the CT images of the head. This information is fed into a detailed mathematical model of how the ultrasound wave spreads through the skull, depending on the angle it strikes.

The mathematical predictions obtained in this way have been found to be in good agreement with measurements taken using a hydrophone (microphone) positioned inside skulls in water (which is used to simulate the brain tissue).

"There is still some way to go before the method can be used on patients, but our focusing technique is the first method to produce a sharp, controlled focus through the skull completely non-invasively and the first that shows a repeatable ability to focus through the skull," says Dr. Clement. "We believe our focusing method and array system together present the first clinically feasible approach to non-invasive trans-skull ultrasound surgery."

(Reference: The research paper, "A noninvasive method for focusing ultrasound through the human skull" by Dr. Greg Clement and Dr. Kullervo Hynynen of Harvard Medical School is being published in Physics in Medicine and Biology Vol 47, issue 8 [21 April 2002] pp 1219-1236.)

[Contact: Dr. Greg Clement, Dr. Kullervo Hynynen]

05-Apr-2002