Theses Doctoral

Optoelectronic Neural Implant Sensors for Cerebral Blood Volume Monitoring

Choi, Christopher Samuel

Nearly 50 million people are afflicted with epilepsy, worldwide. These patients suffer from unprovoked seizures, where neurons in the cerebral cortex under go uncontrolled, hypersynchronous firing of neurons. 30\% of patients with epilepsy do not respond to drug treatments. For these patients, surgical treatment involving the removal or disconnection of brain matter is one of the only alternatives. Such surgical treatments often rely on long-term monitoring of neuronal activity in the brain using subdurally implanted surface electrodes to locate the epileptic focus, but these clinical methods for mapping neuronal activity suffer from low spatial resolutions and poor noise, which can limit the success of surgical treatments where an error of even 1 mm can be critical.
The work described here involves the development of an implantable system for performing optical recordings of intrinsic signal (ORIS) on the surface of the brain. By taking advantage of the unique absorption spectrum of hemoglobin, cerebral blood volume (CBV) can be measured via reflectivity changes in the brain at at specific wavelengths of light. Due to the metabolic demands of the brain, the exaggerated neuronal activity and spiking associated with epileptic seizures can be detected indirectly through changes in CBV. While high resolution ORIS measurements have been recorded using externally mounted CCD sensors, this work presents some of the first developments in producing a fully implantable ORIS sensor.
Progress in the development of an implantable ORIS sensor described here includes: an implantable organic light emitting diode (OLED) and organic photodetector (OPD) integrated on a highly flexible parylene-c substrate, an implantable sensor using a microLED array embedded on a flexible polyimide substrate, and the application of quantum dots to microLEDs for optical down-conversion. Successful in vivo detection of seizures is achieved with high signal-to-noise using these methods. Additionally, spatial localization of seizure activity is performed using the microLED array. These developments represent crucial first steps in the development of a full 2D neuronal mapping system using implantable ORIS devices.


  • thumnail for Choi_columbia_0054D_14605.pdf Choi_columbia_0054D_14605.pdf application/pdf 14.2 MB Download File

More About This Work

Academic Units
Electrical Engineering
Thesis Advisors
Kymissis, Ioannis
Ph.D., Columbia University
Published Here
May 14, 2018