Research‎ > ‎

Group Projects

Our current group in pharmaceutical sciences works in drug development, with a focus in rare diseases and tropical diseases.  We use artificial intelligence and machine learning techniques to find novel compounds to treat these rare diseases.

Rare Diseases

Tropical Diseases

Our group in Electrical and Computer Engineering works on Computational as well as Astrobiological projects.

shredded paper
Computer techniques designed to virtually piece shredded paper back together can be used to piece together Internet cookies to reconstruct an online life.

Drug Development for Dyslipidemia and Atherosclerosis

Natural products provide a basis for a pipeline of drugs that the group is developing at the University of Kentucky College of Pharmacy.  These drugs may one day become important agents for preventing heart attack and stroke.

Accurately Predicting System Success and Failure

A project designed to predict when and how a complex system comprising many nodes might fail  is underway .

Near-IR Imaging

A new near-IR InSb focal plane array video camera is being used to image carotid plaques during carotid endarterectomy. We have just completed a new program that manipulates frames from the camera for analysis was just completed. This new program is entirely menu/mouse driven, and enables users to grab a frame collected at any wavelength just by pointing at it. The mouse can be used to zoom in and out on any selected portion of the image, and to rescale the colors and change color weighting schemes on any area of the image. More importantly, the mouse can be used to select any feature in the image at one wavelength, and the complete spectra (absorbances from all frames) are automatically selected simultaneously and stored in a 2-D variable that can be passed to all of the 90+ programs that we have developed over the past ten years. The completion of this new imaging program will accelerate greatly our analysis of spectra obtained with the near-IR imaging system.

Figure 1 is a near-IR image of a carotid endarterectomy taken at 2312 nm, where -CH3 groups in lipids have an absorbance. The carotid plaque is the most dense red area toward the center of the image. The white circle on left is a white marble placed as a reference in the field of the camera (the black marble also used as an intensity reference is not visible at this wavelength, except for a small highlight to the right of the white marble). The long white features of the image are the tissue retractors used to hold the incision open. The marble and stainless steel instruments appear white in the image because they reflect the most near-IR light. The purple area to the far right is the surgical drape, which reflects most of the near-IR light falling upon it. The color scheme, from lowest light absorbance to highest, is white, violet, blue, green, yellow, orange, and red.

Figure 2 is a zoom-in image of the carotid from Figure 1, shown at the protein amide absorbance wavelength at 2180 nm. The carotid artery appears orange and a red plaque appears at the carotid bifurcation.

Grid Supercomputing

Near-IR spectra obtained at d wavelengths are represented as single points in a d-dimensional hyperspace. The supercomputer calculates the probability that certain regions of the carotid plaques are normal tissue based on their spatially resolved near-IR spectra. If the tissue is abnormal, the direction of the spectral vector of the abnormal tissue in wavelength hyperspace identifies the constituents that make that region abnormal, and the length of the vector (scaled by the probability of a normal spectrum lying in that direction) gives the amount of constituents present. The probabilities calculated on the supercomputer are converted to images on a workstation.


The development of a magnetohydrodynamic acoustic-resonance near-infrared (MAReNIR) spectrometer is currently underway. The MAReNIR spectrometer is a novel device for noninvasive chemical analysis. A major application for the device is near-infrared detection and quantification of cholesterol and lipoproteins simultaneously in serum samples and perhaps even in vivo. Near-IR spectrometry has been shown to be an effective method of determining cholesterol and lipoproteins in a human blood matrix, but variations in sodium ion concentrations constitute a significant interference in results. Knowledge of the ion concentration enables one to overcome the interference to cholesterol, but such knowledge is difficult to obtain noninvasively. The MAReNIR spectrometer overcomes the sodium ion interference by inducing ion motion in a magnetic field with a tunable acoustic wave (see Figure 3). The moving ions create an electrical current that is picked up by electrodes in a cuvette or on the surface of the skin. Measuring the current continuously in a computerized pattern recognition algorithm reveals the ion concentration, and permits accurate analyses of cholesterol. The acoustic wave itself is used to improve identification and quantification of similar apolipoproteins (such as apoA-I and apoA-II) in solution by modulating their conformations and hence their near-IR spectra (through hydrogen bonding). In addition, the acoustic waves help to set the near-IR spectral baseline by establishing the bulk density of tissue samples in vivo.

The MAReNIR spectrometer will be used to create a desktop clinical instrument for analyzing blood for lipoproteins and cholesterol simultaneously and without reagents in physician's offices. Present analytical techniques are error-prone, slow, and expensive. An analytical method that minimizes sample handling and time of analysis will make lipoprotein measurements more accurate by reducing degradation of the sample during the analysis. The spectrometric nature of the instrument may enable it to determine accurately cholesterol and lipoprotein levels directly through the skin as well.


The group is researching carbohydrate compounds for use in the treatment of diabetes.  A unique sweetener that actually controls blood sugar just might be the dream of diabetics everywhere. Dr. Lodder (president of the drug company Spherix) recently completed an interim analysis of the global phase 3 clinical trial results for D-tagatose, a novel compound used in the treatment and management of Type 2 diabetes, that showed a reduction in HbA1c and key blood lipids as well as Body Mass Index (BMI).

OPO Laser

A unique center for near-infrared imaging was constructed around a tunable KTP/OPO laser system granted by the National Institutes of Health to a group of medical researchers at the University of Kentucky. The laser system consists of a MIRAGE 3000B Mid-Infrared Optical Parametric Generator and a Continuum NY81-10 Nd:YAG Pump Laser (see Figure 4). The system provides tunable near-IR light with a wavelength from 1.4 to 4.1 micrometers with an effective power of 3.3 million watts. Figure 5 shows the mirrors in the laser OPO: (a) shows the laser off and (b) shows the laser on. From left to right, the three mirrors are for the 532 nm OPO beam, the 1064 nm OPA beam, and the output coupler, which reflects the Nd:YAG fundamental into a beam dump. Figure 6 shows the KTP optical parametric amplifier crystals that produce output near-IR light (signal and idler beams). Additional hardware and software were consturcted at the University of Kentucky to make the laser more useful; this equiptment is described in the Appendix. A laser control program was constructed and is detailed in a flow chart. The laser detector unit that was constructed is shown in Figure 7. The light is used in medical imaging experiments. The laser system is the first of its kind to be used in near-IR medical research, and allows the production of video images to monitor the progression of disease. These video images contain more specific chemical information on lipids and proteins than is available through CAT or MRI imaging.

The equipment is employed in the collection of data for several multidisciplinary projects aimed at improving human health through state-of-the-art near-infrared/supercomputer techniques. Each project involves the use of near-infrared spectrometry to image noninvasively chemical and structural changes associated with disease. Traditional methods of analysis typically require the removal of tissue samples, comparison between different subjects, or other indirect means of determining progression or effectiveness of treatment of disease. Because the new system permits nondestructive imaging of disease in living subjects, it enables the researchers to monitor disease directly, within individual subjects, which is a superior method of evaluating experimental treatments.

The system is used primarily in projects that are based in three broad components:

In Situ Analysis of Atherosclerotic Plaque
The Stroke Program, which will employ the equipment in the imaging of patient volunteers with carotid atherosclerosis. The laser will be used as a bright, tunable light source to obtain transcutaneous 3-D spatial resolution of plaques, with chemical composition profiles collected at various stages in lesion development. These data will be correlated with results from duplex ultrasound and lipoprotein electrophoresis from Stroke Program projects. This project seeks to investigate factors that contribute to stroke and treatments that can modulate degree of damage that results from stroke.

In Vivo Chemical Analysis of Atherosclerotic Plaques

The in vivo chemical analysis and high-resolution imaging of the structure of atherosclerotic plaques using a near-IR fiber-optic catheter. New imaging algorithms for massively parallel supercomputers and intraarterial fiber-optic cameras are the most recent technological discoveries in this project. The nondestructive chemical analysis of single lesions over time virtually guarantees a new understanding of the mechanisms of lesion formation and growth.
        1)   Near-IR Spectrometric Imaging Design

The laser system has been used to test experimental catheters that are being developed for use in several projects investigating atherosclerosis. The catheters are now being used to monitor disease progression in living animals and their results will eventually be extended to human studies. The catheter's distal reflector tips are constructed of gold plated steel aand are only 450 micrometers wide. Confirmation that individual tips, imported from Germany, meet the necessary focusing and reflectance specifications can be achieved only by testing with the new laser. Use of a distal reflector catheter tip that is out of specifications would move the focal point and cause light loss, which would waste time, research funds, and subjects. Some projects are supported by the National Sciences Foundation and the Kentucky Affiliate of the American Heart Association.

        2)   Catheter Studies

The laser has also been used as the light source in clinical tests of functioning catheters (described above). In these studies, rabbits are maintained on a high-cholesterol diet to create fatty streaks and lesions in their arteries. The catheters are inserted into the femoral artery and advanced to the aortic arch. As the catheters are slowly withdrawn, near-IR chemical analyses of the lesion can be performed. With all other traditional techniques, vascular tissue must be removed from the animal before cholesterol and ox-LDL can be assayed. The advantage of the near-IR technique is that repeated analyses are possible of the same lesion in the same animal during lesion development.