A.M.A. Lorza, H. Ravi, R.C. Philip, J.P. Galons, T.P. Trouard, N.A. Parra, D.D. Von Hoff, W.L. Read, R. Tibes, R.L. Korn, and N. Raghunand,
"Dose-response Assessment by Quantitative MRI in a Phase-1 Clinical Study of the Anti-Cancer Vascular Disrupting Agent Crolibulin"
Scientific Reports 2020;10:14449.
[PDF via Pubmed]
The vascular disrupting agent crolibulin binds to the colchicine binding site and produces anti-vascular and apoptotic effects. In a multisite phase 1 clinical study of crolibulin (NCT00423410), we measured treatment-induced changes in tumor perfusion and water diffusivity (ADC) using dynamic contrast-enhanced MRI (DCE-MRI) and diffusion-weighted MRI (DW-MRI), and computed correlates of crolibulin pharmacokinetics. 11 subjects with advanced solid tumors were imaged by MRI at baseline and 2–3 days post-crolibulin (13–24 mg/m2). ADC maps were computed from DW-MRI. Pre-contrast T1 maps were computed, co-registered with the DCE-MRI series, and maps of area-under-the-gadolinium-concentration-curve-at-90 s (AUC90s) and the Extended Tofts Model parameters ktrans, ve, and vp were calculated. There was a strong correlation between higher plasma drug Cmax and a linear combination of (1) reduction in tumor fraction with AUC90s>15.8 mM s, and, (2) increase in tumor fraction with ve<0.3. A higher plasma drug AUC was correlated with a linear combination of (1) increase in tumor fraction with ADC<1.1×10−3mm2/s, and, (2) increase in tumor fraction with ve<0.3. These findings are suggestive of cell swelling and decreased tumor perfusion 2–3 days post-treatment with crolibulin. The multivariable linear regression models reported here can inform crolibulin dosing in future clinical studies of crolibulin combined with cytotoxic or immune-oncology agents.
@article {Lorza20,
author = {Lorza, Andres M. Arias and Ravi, Harshan and Philip, Rohit C. and Galons, Jean-Philippe and Trouard, Theodore P. and Parra, Nestor A. and Von Hoff, Daniel D. and Read, William L. and Tibes, Raoul and Korn, Ronald L. and Raghunand, Natarajan},
title = {Dose-response Assessment by Quantitative MRI in a Phase-1 Clinical Study of the Anti-Cancer Vascular Disrupting Agent Crolibulin},
journal = {Scientific Reports},
volume = {10},
number = {},
publisher = {},
issn = {},
doi = {https://doi.org/10.1038/s41598-020-71246-w},
pages = {14449},
year = {2020},
}
R.C. Philip, J.J. Rodríguez, M. Niihori, R.H. Francis, J.A. Mudery, J.S. Caskey, E. Krupinski, and A. Jacob,
"Automated High-Throughput Damage Scoring of Zebrafish Lateral Line Hair Cells After Ototoxin Exposure"
Zebrafish 2018;15(2):145-155.
[PDF via Pubmed]
Zebrafish have emerged as a powerful biological system for drug development against hearing loss. Zebrafish hair cells,
contained within neuromasts along the lateral line, can be damaged with exposure to ototoxins, and therefore, pre-exposure
to potentially otoprotective compounds can be a means of identifying promising new drug candidates. Unfortunately, anatomical
assays of hair cell damage are typically low-throughput and labor intensive, requiring trained experts to manually score hair
cell damage in fluorescence or confocal images. To enhance throughput and consistency, our group has developed an automated
damage-scoring algorithm based on machine-learning techniques that produce accurate damage scores, eliminate potential
operator bias, provide more fidelity in determining damage scores that are between two levels, and deliver consistent results
in a fraction of the time required for manual analysis. The system has been validated against trained experts using linear
regression, hypothesis testing, and the Pearson's correlation coefficient. Furthermore, performance has been quantified by
measuring mean absolute error for each image and the time taken to automatically compute damage scores. Coupling automated
analysis of zebrafish hair cell damage to behavioral assays for ototoxicity produces a novel drug discovery platform for
rapid translation of candidate drugs into preclinical mammalian models of hearing loss.
@article {Philip18,
author = {Philip, Rohit C. and Rodriguez, Jeffrey J. and Niihori, Maki and Francis, Ross H. and Mudery, Jordan A. and Caskey, Justin S. and Krupinski, Elizabeth and Jacob, Abraham},
title = {Automated High-Throughput Damage Scoring of Zebrafish Lateral Line Hair Cells After Ototoxin Exposure},
journal = {Zebrafish},
volume = {15},
number = {2},
publisher = {Mary Ann Liebert, Inc.},
issn = {},
doi = {10.1089/zeb.2017.1451},
pages = {145--155},
year = {2018},
}
D.W. Todd, R.C. Philip, M. Niihori, R.A. Ringle, K.R. Coyle, S.F. Zehri, L. Zabala, J.A. Mudery, R.H. Francis, J.J. Rodríguez, and A. Jacob,
"A Fully-Automated High-Throughput Zebrafish Behavioral Ototoxicity Assay,"
Zebrafish 2017;14(4):331-342.
[PDF via Pubmed]
Zebrafish animal models lend themselves to behavioral assays that can facilitate rapid screening of ototoxic,
otoprotective, and otoregenerative drugs. Structurally similar to human inner ear hair cells, the mechanosensory
hair cells on their lateral line allow the zebrafish to sense water flow and orient head-to-current in a behavior
called rheotaxis. This rheotaxis behavior deteriorates in a dose-dependent manner with increased exposure to the
ototoxin cisplatin, thereby establishing itself as an excellent biomarker for anatomic damage to lateral line hair
cells. Building on work by our group and others, we have built a new, fully automated high-throughput behavioral
assay system that uses automated image analysis techniques to quantify rheotaxis behavior. This novel system
consists of a custom-designed swimming apparatus and imaging system consisting of network-controlled Raspberry
Pi microcomputers capturing infrared video. Automated analysis techniques detect individual zebrafish, compute
their orientation, and quantify the rheotaxis behavior of a zebrafish test population, producing a powerful,
high-throughput behavioral assay. Using our fully automated biological assay to test a standardized ototoxic dose
of cisplatin against varying doses of compounds that protect or regenerate hair cells may facilitate rapid
translation of candidate drugs into preclinical mammalian models of hearing loss.
@article {Todd17,
author = {Todd, Douglas W. and Philip, Rohit C. and Niihori, Maki and Ringle, Ryan A. and Coyle, Kelsey R. and Zehri, Sobia F. and Mudery, Jordan A. and Francis, Ross H. and Rodriguez, Jeffrey J. and Jacob, Abraham},
title = {A Fully Automated High-Throughput Zebrafish Behavioral Ototoxicity Assay},
journal = {Zebrafish},
volume = {14},
number = {4},
publisher = {Mary Ann Liebert, Inc.},
issn = {},
doi = {10.1089/zeb.2016.1412},
pages = {331--342},
year = {2017},
}
Conference Papers
R.C. Philip, A. Bilgin, and M.I. Altbach,
"Automated Radial Streaking Artifact Suppression with RGB-STAR,"
2021 International Society for Magnetic Resonance in Medicine (ISMRM),
May. 2021, Online, ISMRM2021-3521.
[PDF via ISMRM]
Streaking artifacts in radial MR imaging due to gradient nonlinearities are suppressed using a beamforming algorithm where region growing image segmentation is used to automatically generate the interference covariance matrix. The performance of the automatic streaking artifact suppression algorithm (RGB-STAR) is compared to algorithms based on coil removal and coil weighting and a beamforming algorithm with manual segmentation.
@INPROCEEDINGS{Philip21,
author={Philip, R.C. and Bilgin, A. and Altbach, M.I.},
booktitle={International Society for Magnetic Resonance in Medicine, 2021},
title={Automated Radial Streaking Artifact Suppression with RGB-STAR},
year={2021},
month={May},
pages={ISMRM2021-3521},
keywords={},
doi={},}
R.C. Philip, A.G. Agosto, N. Sampson, R. Janardhanan, F. Rischard, R.R. Vanderpool, and E. Tubaldi,
"Performance of Supervised Classifiers for Classification of the Four Chambers of the Heart,"
SB3C 2020 Summer Biomechanics, Bioengineering and Biotransport Conference,
Jun. 2020, Vail, Co, SB3C2020-650.
[PDF via SB3C]
Pulmonary hypertension (PH) is clinically defined by a resting
mean pulmonary artery pressure of 25mmHg or greater and is a
heterogenous group of diseases. Pulmonary arterial hypertension
(PAH) is a progressive disease of the pulmonary vasculature leading to
increased pulmonary vascular resistance, elevated pulmonary artery
pressure, right ventricular (RV) dysfunction, and ultimately RV failure
and death. PH related to left heart disease (PH-LHD) is distinct from
PAH and represents the most common form of PH, accounting for 65-
80% of the cases. 5 The proper distinction between PAH and PH-LHD
may be challenging, yet it has direct therapeutic consequences.
Cardiac magnetic resonance (CMR) imaging is used during the
diagnosis and treatment in patients with PH. Studying the shape and
motion of each cardiac chamber during the cardiac cycle in detail allows
for better understanding of PAH and PH-LHD which in turn would
allow better quantification and prediction of patient outcomes. An
intermediate step is the identification of each chamber of the heart.
Currently, expert cardiologists and pulmonologists distinguish between
the right and left ventricles (and the right and left atria) by manually
looking at CMR images of the heart. As the number of patients and the
number of images per patient increase, this process becomes tedious and
is prone to human error. Therefore, a faster screening method using
machine learning to automatically identify the four chambers of the
heart is needed and would allow cardiologists and pulmonologists to
build more accurate models of pulmonary hypertension and heart
disease.
Supervised classifiers such as support vector machines, random
forests, neural networks, etc., are widely used to perform classification
tasks in a wide variety of fields. Using a set of training data with
associated class labels and a list of features describing the training data,
these classifiers automatically learn decision boundaries in a high-
dimensional feature space to separate the training data into the classes
of interest.
In this study, we have developed an algorithm that will
automatically extract features of interest from segmented CMR images
of the four chambers of the heart: right atrium (RA), left atrium (LA),
right ventricle (RV), and left ventricle (LV). We then analyze the
performance of a suite of supervised classifiers using these features of
interest.
The aims of this study are as follows: (1) To segment the four
chambers of the heart; (2) to identify distinct cardiac features that
discriminate between the four chambers of the heart in general and the
RV and LV specifically; and (3) to quantify the performance of a suite
of supervised classifiers to perform the classification of the four
chambers of the heart and the RV and LV.
@INPROCEEDINGS{Philip20,
author={Philip, R.C. and Agosto, A.G. and Sampson, N. and Janardhanan, R. and Rischard, F. and Vanderpool, R.R. and Tubaldi, E.},
booktitle={Summer Biomechanics, Bioengineering and Biotransport Conference, 2020},
title={Performance of Supervised Classifiers for Classification of the Four Chambers of the Heart},
year={2020},
month={June},
pages={SB3C2020-650},
keywords={},
doi={},}
A.G. Agosto, R.C. Philip, R.R. Vanderpool, and E. Tubaldi,
"Ventricular Dynamics in Patients with Pulmonary Arterial Hypertension,"
SB3C 2020 Summer Biomechanics, Bioengineering and Biotransport Conference,
Jun. 2020, Vail, Co, SB3C2020-731.
[PDF via SB3C]
Pulmonary Arterial Hypertension (PAH) is a progressive disease
that is diagnosed in patients with a mean pulmonary artery pressure of
20 mmHg or more, a pulmonary artery wedge pressure of 15 or less,
and a pulmonary vascular resistance of 3 WU or more. Increased
pulmonary artery pressure is the consequence of a decrease in the
radius of the resistance vessels in the lung that results in increased
pulmonary vascular resistance. In response to increased resistance, the
right ventricle can adapt with increased contractility or dilation of the
ventricle. PAH can cause a multitude of symptoms which are
mostly vague, but include, chest pain, bloody cough, shortness of
breath, and weakness. Right ventricular function is a major
determinate of prognosis in patients with PAH.
Cardiac Magnetic Resonance (CMR) imaging is a powerful tool
to assess the volume in the four chambers of the heart (Right and Left
Atrium, and Right and Left Ventricles). A decrease in the ratio of
stroke volume to end-systolic volume is a predictor of mortality in
patients with pulmonary hypertension. Ventricular adaptation to
increased afterload can cause altered ventricular dynamics that has not
been well studied.
The study aims were 1) to segment the right and left ventricles
throughout the cardiac cycle using a custom semi-automatic method
and 2) to investigate the change in right ventricular dynamics during
systole and diastole in patients with pulmonary arterial hypertension.
These novel markers may assist with identification of improved RV
function in patients with PAH.
@INPROCEEDINGS{Agosto20,
author={Agosto, A.G. and Philip, R.C. and Vanderpool, R.R. and Tubaldi, E.},
booktitle={Summer Biomechanics, Bioengineering and Biotransport Conference, 2020},
title={Ventricular Dynamics in Patients with Pulmonary Arterial Hypertension},
year={2020},
month={June},
pages={SB3C2020-731},
keywords={},
doi={},}
R.C. Philip, S.R.S.P. Malladi, M. Niihori, A. Jacob and J.J. Rodríguez,
"Performance of Supervised Classifiers for Damage Scoring of Zebrafish Neuromasts,"
2018 IEEE Southwest Symp. on Image Analysis and Interpretation,
Apr. 2018, Las Vegas, NV, pp. 113-116.
[PDF via IEEEXplore]
Supervised machine learning schemes are widely used to perform classification tasks. There is a wide
variety of classifiers in use today, such as single- and multi-class support vector machines, k-nearest
neighbors, decision trees, random forests, naive Bayes classifiers with or without kernel density
estimation, linear discriminant analysis, quadratic discriminant analysis, and numerous neural network
architectures. Our prior work used high-level shape, intensity, and texture features as predictors in a
single-class support vector machine classifier to classify images of zebrafish neuromasts obtained using
confocal microscopy into four discrete damage classes. Here, we analyze the performance of a multitude
of supervised classifiers in terms of mean absolute error using these high-level features as predictors.
In addition, we also analyze performance while using raw pixel data as predictors.
@INPROCEEDINGS{Philip18,
author={Philip, R.C. and Malladi, S.R.S.P. and Niihori, M. and Rodriguez, J.J.},
booktitle={Image Analysis and Interpretation (SSIAI), 2018 IEEE Southwest Symposium on},
title={A comparison of tracking algorithm performance for objects in wide area imagery},
year={2018},
month={April},
pages={113-116},
keywords={Bayes methods;biology computing;decision trees;feature extraction;image classification;image texture;nearest neighbour methods;neural net architecture;support vector machines;high-level shape;predictors;single-class support vector machine classifier;zebrafish neuromasts;discrete damage classes;supervised classifiers;high-level features;supervised machine learning schemes;multiclass support vector machines;k-nearest neighbors;decision trees;random forests;naive Bayes classifiers;kernel density estimation;linear discriminant analysis;quadratic discriminant analysis;neural network architectures;damage scoring;intensity;texture features;confocal microscopy;image classification;Support vector machines;Biological neural networks;Feature extraction;Backpropagation;Forestry;Neurons;Decision trees;Neural network;support vector machine;random forest;naive Bayes classifier;supervised learning},
doi={10.1109/SSIAI.2018.8470377},}
R.C. Philip, S. Ram, X. Gao and J.J. Rodríguez,
"A Comparison of Tracking Algorithm Performance for Objects in Wide Area Imagery,"
2014 IEEE Southwest Symp. on Image Analysis and Interpretation,
Apr. 2014, San Diego, CA, pp. 109-12.
[PDF via IEEEXplore]
Object tracking is currently one of the most active research areas in computer vision.
In this paper we compare and analyze the performance of six recent object tracking algorithms on a raw,
low resolution, unregistered, interlaced aerial video of multiple cars moving on a roadway.
This dataset comprising 50 frames of video offers a wide variety of challenges related to imaging issues
such as low resolution, unregistered frames, camera motion, and interlaced video, as well as object detection
problems such as low contrast, background clutter, object occlusion and varying degrees of motion.
We present the performance of these algorithms in terms of both overlap accuracy and the Euclidean distance
of the center pixel returned by the tracking algorithm from the ground truth.
@INPROCEEDINGS{Philip14,
author={Philip, R.C. and Ram, S. and Gao, X. and Rodriguez, J.J.},
booktitle={Image Analysis and Interpretation (SSIAI), 2014 IEEE Southwest Symposium on},
title={A comparison of tracking algorithm performance for objects in wide area imagery},
year={2014},
month={April},
pages={109-112},
keywords={computer vision;image registration;image resolution;object tracking;road vehicles;traffic engineering computing;video signal processing;Euclidean distance;computer vision;interlaced aerial video;low resolution video;multiple cars;object tracking algorithm;roadway;unregistered video;wide area imagery;Accuracy;Clutter;Computer vision;Image resolution;Object tracking;Robustness;Object tracking;localization error;overlap area;partial occlusion;wide area imagery},
doi={10.1109/SSIAI.2014.6806041},}
R.C. Philip, J. J. Rodríguez, and R.J. Gillies,
"Seed Pruning using a Multi-Resolution Approach for Automated Segmentation of Breast Cancer Tissue,"
2008 IEEE Intl. Conf. on Image Processing,
Oct. 2008, San Diego, CA, pp. 1436-9.
[PDF via IEEEXplore]
This paper proposes a new automated system for segmentation of breast cancer tissue.
The segmentation algorithm involves a principal component region growing scheme for high-resolution images.
The number of candidate seed pixels is extremely large due to the high resolution.
The main focus of this paper is to present a multi-resolution scheme for accurate selection of seed pixels
to be presented as inputs to the region growing segmentation algorithm.
The system is tested for accuracy, and the efficiency is measured in terms of percentage reduction in
number of seed pixels, as well as accuracy of the segmentation results.
@INPROCEEDINGS{Philip08,
author={Philip, R.C. and Rodríguez, J.J. and Gillies, R.J.},
booktitle={Image Processing, 2008. ICIP 2008. 15th IEEE International Conference on},
title={Seed pruning using a multi-resolution approach for automated segmentation of breast cancer tissue},
year={2008},
month={Oct},
pages={1436-1439},
keywords={biological tissues;cancer;image resolution;image segmentation;medical image processing;principal component analysis;automated segmentation;breast cancer tissue segmentation;multiresolution approach;principal component region growing scheme;region growing segmentation algorithm;seed pruning;Breast cancer;KL transform;region growing;seed pixels},
doi={10.1109/ICIP.2008.4712035},
ISSN={1522-4880},}
PhD Dissertation
R.C. Philip,
"Generalized Performance Measures for Object Detection,"
[PDF via UA Libraries]
Classical detection theory has long used traditional measures such as precision, recall, F measure, and G measure to evaluate the quality of detection results. Such evaluation can be done for performance analysis of competing detection algorithms, or for parameter tuning to optimize parameters based on training data. This performance analysis can be done at the pixel level or at the object level. Conventional performance measures to quantify detection accuracy against known ground truth are effective when applied at the pixel level or when applied at the object level with simple detection outcomes. In many cases, however, object-level detection often results in hybrid detections such as a single ground truth object split into multiple detected objects (i.e., split detections) or multiple ground truth objects merged into a single detected object (i.e., merged truths) and combinations thereof. In such cases, conventional performance measures are ineffective. A new generalized framework for evaluating object detection algorithms is proposed. This generalized framework introduces two new precision measures and two new recall measures, resulting in four new F and G measures, all of which reduce to their classical detection theory counterparts in cases with simple detection outcomes (no split detections or merged truths). The new concept of shared positives is developed and the shared positive (SP) curve is proposed as a performance evaluation tool. The performance analysis of the new generalized framework includes evaluation of eight object detection algorithms. Two of these new generalized F measures perform better than the classical F measure during parameter tuning and algorithm performance comparison, while a third offers comparable performance. One of these new generalized F measures can serve as a replacement for classical F measure to perform object detection evaluation more accurately. In addition, we also develop three high-throughput zebrafish ototoxicity assays: a) an anatomical assay which uses feature extraction techniques and machine learning to automatically quantify damage to zebrafish neuromasts; b) a behavioral assay which uses object detection and orientation computation to automatically measure rheotaxis index; and c) a novel tracking assay which uses optical flow to track zebrafish in video data and provide a comprehensive analysis of zebrafish swimming behavior. The behavioral assay serves as a test case to demonstrate the effectiveness of the new generalized F measures for zebrafish detection algorithm tuning.
@PHDTHESIS{Philip08,
author={Philip, R.C.},
title={Generalized Performance Measures for Object Detection},
school={University of Arizona},
year={2020},}
Masters Thesis
R.C. Philip,
"Seed Pruning for Multi-Resolution Segmentation of Vasculature in Immunohistochemical Images,"
[PDF]
An automated system for the segmentation of vasculature in high-resolution digital color images
of immunohistochemically stained slides of breast cancer tissue is presented.
Manual immunohistochemical staining processes result in differences in stain retention by the
regions of interest (ROIs).
Difficulties in the segmentation processes arise when the stain retention is poor,
causing the existing color profiling schemes in the literature to fail.
The automated segmentation algorithm uses a principal component region-growing scheme with
automated generation of seed pixels.
The number of candidate seed pixels is extremely large due to the high-resolution.
The main focus of the research is to present a multi-resolution scheme for accurate selection
of a minimal number of seed pixels to be presented as inputs to the region-growing segmentation
algorithm.
The system is tested for accuracy, and the efficiency is measured in terms of
percentage reduction in number of seed pixels, as well as accuracy of the segmentation
results and false positive and false negative pixel classification ratios.
The effect of the number of resolution levels on the accuracy and number of seed pixels
selected is also discussed.
@MASTERTHESIS{Philip08,
author={Philip, R.C.},
title={Seed Pruning for Multi-Resolution Segmentation of Vasculature in Immunohistochemical Images},
school={University of Arizona},
year={2008},}
Under Review/In Preparation
R. C. Philip, and J.J. Rodríguez,
"Generalized Precision, Recall, and F Scores for Evaluation of Object Detection Performance."
(In preparation).
[PDF Placeholder]
[BibTeX Placeholder]
Abstract—Classical detection theory has long used traditional measures such as precision, recall, F measure, and G measure to
evaluate the quality of detection results. Such evaluation can be done for performance analysis of competing detection algorithms, or
for parameter tuning to optimize parameters based on training data. The performance analysis can be done at the sample level (e.g.,
pixel-by-pixel) or at the object level (e.g., a set of connected pixels are grouped together into one object). Conventional performance
measures are effective when applied at the sample level or when applied at the object level with simple detection outcomes. In many
cases, however, object-level detection often results in hybrid detections such as a single ground-truth object split into multiple detected
objects or multiple ground-truth objects merged into a single detected object and combinations thereof. In such cases, the conventional
performance measures are ineffective. We propose a generalized framework for evaluating object detection algorithms, which involves
two new precision measures and two new recall measures, resulting in four new F and G measures, all of which reduce to their
classical detection theory counterparts in cases with simple detection outcomes (no split detections or merged truths). We introduce
the shared positive detection vs. shared positive truth (SPD-SPT) curve, and evaluate the detection performance of eight object
detection algorithms using our generalized framework.
In preparation. To be updated soon.
R. C. Philip, J. J. Rodríguez, D.W. Todd, M. Niihori, L. Zabala, and A. Jacob,
"Complementing Rheotaxis Index with Additional Behavioral Metrics by Tracking Zebrafish in High-Throughput Behavioral Ototoxicity Assays"
(In preparation).
[PDF Placeholder]
[BibTeX Placeholder]
Rheotaxis index is well established as a biomarker for anatomic damage to lateral line hair cells,
and has been widely used in several zebrafish behavioral assays to facilitate rapid screening of
ototoxic, otoprotective, and otoregenerative drugs. Automation of rheotaxis index quantification
has shortened experiment time by several orders of magnitude resulting in high-throughput state-
of-the- art behavioral assays. These fully automated zebrafish behavioral assays rely on image
processing software that detect zebrafish from frames of video data, compute their orientation
with respect to a flow of water, and determine the rheotaxis index of a test population. A
limitation of simply automating rheotaxis index computation is the inability to distinguish
between fish that are actively swimming from ones that are either stuck to the bottom of the tank
or being pushed back by the current. We improve upon existing zebrafish behavioral assays by
tracking each individual zebrafish from frame to frame throughout the experiment. In addition,
we develop additional behavioral metrics that quantify swimming activity such as distance
traveled by the zebrafish, turning activity, periods of rest, etc. These novel behavioral metrics
complement rheotaxis index and help identify zebrafish that actively exhibit rheotaxis swimming
behavior strengthening the ability to screen for ototoxicity.
In preparation. To be updated soon.
Posters
D.W. Todd, R. C. Philip, M. Niihori, J.J. Rodríguez, and A. Jacob.
"High-Throughput Behavioral Zebrafish Assay for Drug Development Targeting Hearing Loss,"
(abstract) AOS 149th Annual Meeting, American Otological Society, Chicago, IL, May 20-21, 2016. Best Poster Award.