Optical techniques are widely used for the imaging of neural activity in the brain, but conventional microscopy approaches (e.g. wide field, confocal, two-photon, etc.) are limited in the depth of tissue that can be imaged. A different class of techniques utilizes implantable optical waveguides, or endoscopic-like probes, to efficiently convey light from deep tissue to an optical system positioned external to tissue. While these techniques have proven highly successful with a number of commercial products, their complexity in terms of optical components is high, their use causes significant tissue damage, and their field-of-view is narrow. Normally, achieving a wider field-of-view requires a wider endoscopic probe, which leads to increased tissue damage. Our work breaks some of the limitations of contemporary endoscopic imaging by [1]:  (1) applying machine learning to reduce the hardware complexity;  and (2) modifying the tip of the endoscopic probe to increase the field-of-view in image space while maintaining narrow probe diameter.

The student will use state-of-the-art optical design and analysis software to explore and optimize modifications to the endoscopic probe tip.  They will also evaluate the effects of these modifications on the acceptance cone angle of the probe. Once optimized, the student will generate a set of synthetic datasets to use in training a deep neural network to learn the mapping between the light pattern at the proximal end of the probe based upon neural object(s) located near the distal end. Further work can be performed in testing this trained neural network on de novo neural circuits.  Example questions to be answered include: (1) Can the field of view be increased without creating gaps in the angular field? (2) Can multiple field depths be addressed simultaneously? The desired outcome of this project is a systematic study that addresses these scientific questions.

[1]       R. Guo et al., “Overcoming the field-of-view to diameter trade-off in microendoscopy via computational optrode-array microscopy,” Optics Express, vol. 31, no. 5, pp. 7705-7514, 2023, doi: https://doi.org/10.1364/OE.478314.

The goal of Prof. Furse’s project is to provide a new measurement modality for reflections and transmissions in microwave medical imaging applications [1, 2]. This would enable measurements that are much faster (seconds rather than minutes) and help solve a major problem for microwave imaging where today’s measurements take too long, the subject moves/breathes, and the image blurs. Ideally, this system would also be smaller, more robust, and less expensive than today’s technologies – potentially even inexpensive enough to be disposable.

To achieve the goals of this project, the equipment needs to be converted from taking one-port reflectometry impedance measurements to also taking two-port transmission measurements. As such, the student will learn both hardware and software concepts, and how they integrate to give the final parameters of the system. Example questions to be answered include: (1) What is needed to accurately synchronize or calibrate two-port transmission measurements in the time domain? (2) What is needed to accurately convert these to and from the frequency domain?  The desired outcome of this project are methods and assessments of reflection/transmission measurement and software for detecting changes in the breast.

[1]       J. B. H. E. Benoit, and C. M. Furse, “Evaluation of Impedance Measurement Using Spread Spectrum Time Domain Reflectometry,” IEEE Transactions on Instrumentation and Measurement, vol. 72, pp. 1-8, 2023, doi: doi: 10.1109/TIM.2023.3295463.

[2]       E. Porter, H. Bahrami, A. Santorelli, G. Gosselin, L. A. Rusch, and M. Popović, “A wearable microwave antenna array for time-domain breast tumor screening,” IEEE transactions on medical imaging vol. 35, no. 6, pp. 1501-1509, 2016.

The unique electronic and optoelectronic properties of carbon nanotubes (CNTs) [1] have made them a promising material platform for neuromorphic devices and systems, such as synaptic transistors in nanoelectronic crossbars [2, 3] and phototransistors for smart machine visions [4]. The key is to achieve wafer-scale aligned films of single chirality CNTs, which preserve nanoscale 1D properties at the macroscale. Dr. Gao has pioneered a solution-based self-assembly vacuum filtration technique to prepare such films, but their alignment quality is suboptimal [5].

The goal of this project is to explore multiple process parameters of controlled vacuum filtration to further improve the alignment quality. Specifically, the student will systemically investigate how CNT colloidal properties and filter membrane properties affect the degree of alignment of obtained films. The student will also learn how to operate horn sonicators to control the CNT length in aqueous suspensions and how to characterize prepared samples using optical methods. In addition, the students will construct a simple contact angle measurement setup to characterize the hydrophilicity of filter membranes purchased from different manufacturers. Finally, they will explore to develop explanations for the experimental observations.  For example, what are the parameters and how do these parameters dictate the self-assembly process of CNTs?  The desired outcome of this project is to advance manufacturing processes to achieve the holy-grail wafer-scale films of crystalline CNTs.

[1]       A. Jorio, Gene Dresselhaus, and Mildred Dresselhaus, Carbon nanotubes: advanced topics in the synthesis, structure, properties and applications. Berlin: Springer, 2008.

[2]       S. Kim, Bongsik Choi, Meehyun Lim, Jinsu Yoon, Juhee Lee, Hee-Dong Kim, and Sung-Jin Choi, “Pattern recognition using carbon nanotube synaptic transistors with an adjustable weight update protocol,” Acs Nano, vol. 11, no. 3, pp. 2814-2822, 2017.

[3]       I. Sanchez Esqueda, Xiaodong Yan, Chris Rutherglen, Alex Kane, Tyler Cain, Phil Marsh, Qingzhou Liu, Kosmas Galatsis, Han Wang, and Chongwu Zhou, “Aligned carbon nanotube synaptic transistors for large-scale neuromorphic computing,” Acs Nano, vol. 12, no. 7, pp. 7352-7361, 2018.

[4]       Q. B. Zhu, Li, B., Yang, D.D., Liu, C., Feng, S., Chen, M.L., Sun, Y., Tian, Y.N., Su, X., Wang, X.M. and Qiu, S., “A flexible ultrasensitive optoelectronic sensor array for neuromorphic vision systems,” Nat Commun, vol. 12, no. 1, pp. 1-7, 2021.

[5]       F. Katsutani et al., “Direct observation of cross-polarized excitons in aligned single-chirality single-wall carbon nanotubes,” (in English), Phys Rev B, vol. 99, no. 3, Jan 16 2019, doi: ARTN 035426 10.1103/PhysRevB.99.035426.

Our work aims to understand what can be observed in challenging environments such as fog, snowstorms, sandstorms, and low-light conditions. We investigate how properties of the electromagnetic field—like polarization, coherence, and spectrum—can reveal additional scene information. This leads to questions about co-optimizing photonic hardware and software, including machine learning, to enhance observation and detection.

Through simulations and experiments, we explore both scientific questions and practical applications, such as improving autonomous vehicle performance across land, sea, air, and space, and investigating biological processes like temperature effects on neural activity. We also develop novel sensors for spectro-polarimetric astronomical  imaging.

Our lab encourages cross-disciplinary collaboration and has produced multiple publications led by REU students, including work on machine-learning-enabled defogging1 and predicting thermal images from RGB sensors.2

The student will run simulations and perform experiments to evaluate optical systems under various lighting conditions. If time allows, outdoor measurements will be conducted. The student will have access to a supercomputing cluster (TACC) and a fully equipped optics lab. For more details on recent publications, visit https://www.rajeshmenon.net

The goal of this project is to enhance the capabilities of neural networks to function robustly and flexibly in real-world settings with a focus on machine vision applications. Recent advances in machine learning and hardware have created artificial neural networks (ANNs) that have achieved high accuracies in some applications, e.g., image classification. In spite of these advances, such systems cannot match the performance, robustness, flexibility, or energy efficiency of the human brain. It has been hypothesized that this gap is due to the powerful properties of the brain as a computing machine. To bridge this gap, recent research has created simplified models of biological neurons and used them to create spiking neural networks (SNNs) [1, 2].

Advancing the computational capabilities of SNNs necessitates new signal encoding strategies and learning algorithms that can incorporate the different brain’s computational and circuit principles, such as high parallelism, compression, mixed digital and analog signal processing and communication, nonlinearity, or stochasticity. The students will investigate machine vision applications, which exploit the properties of neuro-inspired SNNs and perform basic comparison studies of these SNNs and other ANN systems for artificial vision applications. Example questions to be answered include: (1) What are the architectural and algorithmic bottlenecks in existing SNNs for energy efficiency, and computational robustness and flexibility? (2) How does incorporating the brain inspired temporal coding paradigm impact the real-time processing capabilities of the new SNNs?  The desired outcome of this project is SNN solutions for machine vision, with superior real-time processing and learning capabilities of ANNs by exploiting the benefits of brain-like temporal codes and mixed digital and analog processing. The student are expected to have basic knowledge of machine learning and some experience of how to train simple neural network architectures.

[1]        B. V. Benjamin, Peiran Gao, Emmett McQuinn, Swadesh Choudhary, Anand R. Chandrasekaran, Jean-Marie Bussat, Rodrigo Alvarez-Icaza, John V. Arthur, Paul A. Merolla, and Kwabena Boahen, “Neurogrid: A mixed-analog-digital multichip system for large-scale neural simulations,” Proceedings of the IEEE, vol. 102, no. 5, pp. 699-716, 2014.

[2]        L. Khacef, Nassim Abderrahmane, and Benoit Miramond, “Confronting machine-learning with neuroscience for neuromorphic architectures design,” 2018 International Joint Conference on Neural Networks (IJCNN), pp. 1-8, 2018.

The goal of this research is to design a Deep Neural Network (DNN)-assisted 300 GHz long-range outdoor communication link with a data rate of 100 Gbps, capable of rapidly optimizing its performance in response to field variations. With the growth of smart cities, the demand for robust, ultra-high-speed communication links is increasing drastically. However, the limited bandwidth available at RF and mm-wave frequencies cannot keep pace with this growth, even with employing sophisticated modulation techniques. On the other hand, the terahertz range offers a vast, unallocated spectrum that can provide the ultra-wide bandwidth necessary for high-speed communications. Recent researchers have demonstrated the potential of integrated circuit technologies to implement 300 GHz communication links with data rates exceeding 100 Gbps. Nevertheless, numerous unresolved challenges remain, which can be classified into three main categories: 1) integrated circuit design, 2) high-gain antenna design, and 3) the sensitivity of terahertz data links to the environmental conditions. In this project, we aim to address these challenges using DNNs and other optimization methods, tackling hurdles that conventional solutions cannot overcome.

The student will learn how to build DNN models using Python and apply them on terahertz data links under various environmental conditions. Moreover, the student will get familiar with Electronic Design Automation (EDA) concepts in order to learn how to optimize THz circuits using Artificial Intelligence (AI) models.

Implantable electronic medical devices have transformed diagnostics and treatments of many disorders and diseases. These electronic devices must communicate with external equipment for data acquisition and signal generation. Implantable antenna systems are a good option for wireless data and power transfer. However, next-generation implantable medical devices (IMDs) are becoming much smaller and create a major challenge for the antenna design. This project aims to create a new implantable antenna system that can be injected into the body, where body heat transforms it into a soft, conductive antenna. Once injected, the antenna will remotely couple with the much smaller antenna in the implantable medical device without physical attachment. This approach will enable a new class of wireless biotelemetry antennas and their associated medical applications. This novel injectable antenna system will be based on a thermoresponsive nanocomposite created in the Department of Chemical Engineering, with whom we will collaborate. This summer we will focus on the design of the antenna (simulation), considering the variable tissue properties in the body, and its testing and evaluation in ground pork models (measurement).

Contact: Prof. Cynthia Furse cfurse@ece.utah.edu

Recent trends in photonics have shown that optimized pixelated surfaces can outperform classical designs in reducing design area and improving the frequency response. Such pixelated surfaces, on a microstrip implementation, can also be used to realize “random” engineered RF components (e.g., filters, matching networks, power splitters/combiners, etc.). Since there are no closed-form solutions to describe the design of these surfaces, computational design is key. We have recently proposed a method for synthesizing a pixelated surface to act as a filter using a direct binary search and realized based on it a dual-band RF filter [1].  The method can be used to reliably create designs for generalizable RF functions, many of which are yet to be explored.

For this project, the student will optimize device geometries to realize RF surfaces with various additional electromagnetic responses, in particular, power splitters and combiners with tailored frequency responses. An important question is whether inverse-designed structures can attain similar performances as traditional structures, such as a Wilkinson power divider, while simultaneously having smaller area and not requiring the introduction of e.g. a lumped resistor (by means of the structure radiating so as to provide isolation between outputs). Machine learning will be used [2] in order to generate multiple devices with tailored responses from a unique training set. An important second question is what size datasets are required for good component performance.  The student will learn how to build and interface a custom algorithm with a commercial EM solver, how to fabricate selected microstrip circuit designs, and perform tests using vector network analyzers.  The desired outcome of this research will be a working prototype that could be used as a building block for miniaturized microwave circuits.

[1]        J. Lee, Jia, W., Sensale-Rodriguez, B. and Walling, J.S., “Pixelated RF: Random Metasurface Based Electromagnetic Filters,” presented at the 2023 21st IEEE Interregional NEWCAS Conference (NEWCAS), 2023.

[2]        S. Banerji, A. Majumder, A. Hamrick*, R. Menon, and B. Sensale-Rodriguez, “Ultra-compact integrated photonic devices enabled by machine learning and digital metamaterials,” (in English), Osa Continuum, vol. 4, no. 2, pp. 602-607, Feb 15 2021, doi: 10.1364/Osac.417729.

The goal of Prof. George’s project is to provide amputees with intuitive control of, and natural sensory feedback from, dexterous multi-articulate bionic arms. State-of-the-art upper-limb prostheses have become capable of mimicking many of the movements and grip patterns of endogenous human hands. Although these devices have the capabilities to replace much of the motor function lost after hand amputation, the methods for controlling and receiving feedback from these prosthetic limbs are still primitive. In the United States alone, 1 in every 200 individuals suffer from limb loss [1], and up to 50% of amputees abandon their prostheses due to ineffective control and/or a lack of feedback [2]. Utah Slanted Electrode Arrays implanted into the residual arm nerves can provide bi-directional communication between the human user and the bionic arm. The challenge is developing user-specific algorithms for reading/writing electrical information to/from the nervous system that maximize dexterity.

To achieve the goals of this project, better algorithms and more precise training data for the user-specific algorithms are needed. As such, the student will learn how to develop machine learning models using Python and MATLAB, and they will use those models to analyze electrophysiological data. They will also learn about underlying assumptions made when labelling training data, and how to minimize errors in training data.  Example questions to be answered by the project include: (1) What features of neural data contain the most predictive power for kinematic motion? (2) What role does the temporal pattern of neural action potentials play in conveying contact/texture information?  The desired outcome of this project is a list of the key components of neural data that encode sensorimotor dexterity, such that future neuroprostheses could leverage these components to enhance prosthesis adoption and patient outcomes.

[1]       E. A. B. a. T. T. Chau, “Upper limb prosthesis use and abandonment: A survey of the last 25 years,” Prosthet. Orthot. Int., vol. 31, 3, pp. 236-257, 2007, doi: doi: 10.1080/03093640600994581.

[2]       E. B. a. T. Chau, “Upper-limb prosthetics: critical factors in device abandonment,” Am. J. Phys. Med. Rehabil, vol. 86, no. 12, pp. 977-987, 2007, doi: doi: 10.1097/PHM.0b013e3181587f6c.

Measurement of blood pressure (BP) is essential for early diagnosis and management of hypertension, a condition that 45% of US adults have and a risk factor for development of heart failure and the leading cause of death in the U.S.  Frequent out-of-clinic BP measurements are better predictors of cardiovascular events, however, existing technologies requires costly and cumbersome instrumentation setups that prevent their use outside of the clinic or the lab. Accurate out-of-clinic BP measurements still require the use of automated inflation cuffs, but the inflation of the cuff during measurement is known to trigger an increase in BP, and BP measurements during the night cause sleep arousal and stress, leading to higher BP readings. Further, regular adult size cuffs do not fit all subjects which can result in misleading BP readouts. Given the mortality burden of heart failure, rising healthcare costs, and the aging population, technological improvements in continuous BP measurement for prevention and surveillance are crucial.

The goal of Dr. Sanchez Terrones’ project is to develop and evaluate a bioimpedance skin patch that applies electrical current to the body to obtain quantitative hemodynamics measures of BP [1, 2]. The role of the student will consist of designing and optimizing electrodes’ characteristics for blood pressure measurements with electromagnetohydrodynamic (EMHD) simulations with simplified models first, and then with realistic human computable phantoms. The student will learn advanced data modeling techniques and state-of-the-art physics-informed neural networks.  The students will also learn to organize and conduct human oriented research and subject screening.  Example questions to be answered include:  (1) What electrode configuration is optimal and most sensitive to pulsatile blood at the radial artery? (2) What is the optimal electrode material in terms of contact impedance for bioimpedance measurements? The expected outcome of this project is a guideline of best location and electrodes to enable accurate wrist bioimpedance measurements for cuffless blood pressure monitoring.

[1]       G. B. Ha, B. A. Steinberg, R. Freedman, A. Bayes-Genis, and B. Sanchez, “Safety evaluation of smart scales, smart watches, and smart rings with bioimpedance technology shows evidence of potential interference in cardiac implantable electronic devices,” Heart Rhythm, vol. 20, no. 4, pp. 561-571, 2023, doi: DOI:https://doi.org/10.1016/j.hrthm.2022.11.026.

[2]       L. Vela et al., “IoMT-enabled stress monitoring in a virtual reality environment and at home,” IEEE Internet of Things Journal, vol. 10, no. 12, pp. 10649-10661, 2023.

A traumatic brain injury as well as the disease hydrocephalus can cause life-threatening levels of intracranial pressure. It is important to measure this intracranial pressure, which typically involves invasive procedures in a clinical environment.  This adds additional strain on the patient [1]. In this project, a pressure sensor is being developed that is entirely passive and can remain in contact with the cerebrospinal fluid over extended periods of time as part of an implant [2]. This sensor is MRI compatible due to the absence of metals and can be read out by a handheld ultrasound device outside of a clinical setting, thereby improving point-of-care.

The objectives of this project include improving the manufacture of these pressure sensors and a study of their longevity. The longevity tests take place under accelerated conditions due to the project’s time constraints. The student will learn microfabrication techniques and how to evaluate manufactured sensors based on calibration measurements. The student will also learn about reactive accelerating aging and how to interpret the resulting data.  Example questions to be answered include: (1) What are the decisive fabrication features that affect the sensor’s pressure sensitivity?  (2) How long can the pressure sensor reliably measure the intracranial pressure until it must be replaced or re-calibrated?  The desired outcome of this project is an optimized fabrication procedure for pressure sensors with improved sensitivity in the physiological relevant pressure range, as well as strategies on how to extend the implant’s longevity.

[1]       J. B. P. H. Raboel, M. Andresen, B. M. Bellander, B. Romner, “Intracranial Pressure Monitoring: Invasive versus Non-Invasive Methods—A Review,” Critical Care Res Pract., 2012, doi: doi: 10.1155/2012/950393.

[2]       S. B. T. Ahmad, M. Leber, T. J. Garrett, C. F. Reiche, F. Solzbacher, “Fabrication of Polymer Membrane-Suspended Microstructures on Printed Circuit Boards,” Journal of Microelectromechanical Systems, vol. 31, no. 3, 2022, doi: doi: 10.1109/JMEMS.2022.3162785.