A Caltech Library Service

Low-Power and Miniaturized Medical Electronics for In-Vivo Localization and Tracking


Sharma, Saransh (2023) Low-Power and Miniaturized Medical Electronics for In-Vivo Localization and Tracking. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/xrw0-k789.


Medical electronic devices are an integral part of the healthcare system today. Significant advances have been made over the past few decades to yield highly miniaturized and low-power medical devices that are suitable for implantable, ingestible, or wearable applications. A key feature of medical devices that is central to their use in many applications is the capability to locate them precisely inside the body, and quite a lot of research effort has been expended in this direction. Location sensing is crucial for several applications: tracking pills in the GI tract, navigation during precision surgeries, endovascular procedures, robotic and minimally invasive surgery, and targeted therapy. The current gold-standard solutions for these procedures include invasive techniques such as endoscopy, or procedures that require repeated use of potentially harmful X-ray radiation such as CT scans. These techniques also require repeated evaluation in a hospital setting and are not conducive for non-clinical environments. While there are several alternative non-ionizing methods for imaging and localization based on electromagnetic tracking, radio-frequency, ultrasound, and optical tracking, none of them are able to simultaneously achieve a high field-of-view of tracking, high spatiotemporal resolution, fully wireless operation and miniaturization of the sensing devices, and system scalability with the number of devices. In this dissertation, we present a radiation-free system for high-precision localization and tracking of miniaturized wireless devices in vivo, using harmless magnetic field gradients.

First, we demonstrate our system for precision surgery applications. We designed highly miniaturized, wireless and battery-less microdevices, capable of measuring and transmitting their local magnetic field. One such device can be attached to an implant inside the body and another to a surgical tool, such that both can simultaneously measure and communicate the magnetic field at their respective locations to an external receiver. The relative location of the two devices on a real-time display can enable precise surgical navigation without using X-ray fluoroscopy. The prototype device consists of a micro-chip fabricated in 65nm CMOS technology, a 3D magnetic sensor and an inductor-coil. The chip performs wireless power management, wireless bi-directional data-telemetry, and I2C communication with the sensor. Planar electromagnetic coils are designed for creating monotonically varying magnetic fields in the X, Y, and Z directions, resulting in field gradients that encode each spatial point with a unique magnetic field value. The concept of gradient-based spatial encoding is inspired by MRI. The system is tested in vitro to demonstrate a localization accuracy of <100µm in 3D, the highest reported to the best of our knowledge.

Second, we demonstrate our system for localization and tracking of ingestible microdevices in the GI tract, which is valuable for the diagnosis and treatment of GI disorders. We designed highly miniaturized, low-power, and wireless ingestible devices to sense and transmit their local magnetic field as they travel through the GI tract. These devices consist of a 3D magnetic sensor, a Bluetooth microprocessor and a 2.4GHz Bluetooth antenna for wireless communication, all packaged into a 000-size capsule. The magnetic field sensed by the devices is created by using high-efficiency planar electromagnetic coils that encode each spatial point with a distinct magnetic field magnitude, allowing us to track the location of the devices unambiguously. The system functionality is demonstrated in vivo in large animals under different chronic conditions and disease models to show 3D localization and tracking in real time and in non-clinical settings, with mm-scale spatial resolution, and without using any X-ray radiation. This has the potential for significant clinical benefit for quantitative assessment of GI transit-time, motility disorders, constipation, incontinence, medication adherence monitoring, anatomic targeting for drug delivery, and targeted stimulation therapy.

Third, in order to further miniaturize the devices developed for the above two applications and to make them even more low-power, we present a monolithic 3D magnetic sensor in 65nm CMOS technology that measures <5mm² in area and consumes 14.8µW in power while achieving <10μTrms noise. Our novel 3D magnetic sensor overcomes the challenges faced by traditional magnetic sensors by being fully CMOS compatible and achieving high sensitivity with only µW-level power, which is in sharp contrast with Hall and Fluxgate sensors. The sensor is comprised of three orthogonal and highly dense metal coils implemented in the 65nm node, which generate a voltage signal in response to AC magnetic fields by electromagnetic induction. The EMF voltage signal is processed by on-chip circuitry that performs low-noise amplification, filtering, peak detection, and 12-bit digitization. Though the sensor can be used for a variety of applications that require AC field sensing, it is particularly useful for biomedical applications—tracking catheters and guidewires during endovascular procedures, minimally invasive surgeries, targeted radiotherapy, and for use as fiducial markers during preoperative planning. The proposed magnetic sensor is demonstrated for use in 3D tracking of catheters using the magnetic-field gradient-based spatial encoding scheme, and achieves 500µm of mean 3D localization accuracy.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Electrical Engineering, Biomedical Engineering, Integrated Circuits, VLSI, Biodevices
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Electrical Engineering
Awards:Demetriades-Tsafka-Kokkalis Prize in Biotechnology or Related Fields, 2023.
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Emami, Azita
Thesis Committee:
  • Shapiro, Mikhail G. (chair)
  • Emami, Azita
  • Vaidyanathan, P. P.
  • Marandi, Alireza
  • Traverso, Giovanni
Defense Date:15 May 2023
Non-Caltech Author Email:sharma.saransh41 (AT)
Funding AgencyGrant Number
Rothenberg Innovation Initiative (RI2)101170
Heritage Medical Research Institute150901
Record Number:CaltechTHESIS:05312023-034316442
Persistent URL:
Related URLs:
URLURL TypeDescription adapted for Ch. 3 adapted for Ch. 3 adapted for Ch. 4 adapted for Ch. 5
Sharma, Saransh0000-0002-5052-4932
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:15238
Deposited By: Saransh Sharma
Deposited On:01 Jun 2023 16:54
Last Modified:16 Jun 2023 16:46

Thesis Files

[img] PDF - Final Version
See Usage Policy.


Repository Staff Only: item control page