Empowering Precision: Monitoring Nystagmus for Neurological Excellence

Chiropractic adjustments aimed at correcting spinal alignment and relieving muscle tension could potentially alleviate associated symptoms like neck pain or headaches.

A revolution in healthcare
in just a few keystrokes.

C. Light challenges you to leverage Gen AI to medical code and improve image quality for a better early prediction of disease.

Track A

Deep Generative Models

for Image Denoising 

Task: Develop a generative approach to effectively denoise and improve the imaging quality.

Context: Generative learning frameworks — such as Generative adversarial networks (GANs), variational autoencoders (VAEs), Diffusion Models — have been experimented in the past years to learn image embedding. This method minimizes noise while preserving essential details and structures in the image.

Track B

Generative AI for

Medical Coding 

Task: Utilize Gen AI models to learn the latent embedding for each one of these video, text, and numerical pairings and generate synthetic retinal video sequences with correlated eye motions.

Context: The goal is to initiate the generation of synthetic retinal video sequences seamlessly incorporating correlated, specific eye motions. Available data includes retinal video sequences along with associated data—i.e. eye motion trace, diagnosed disease state, and meta-data such as eye motion and its statistics.

Track X

Your Original Solution 

using Gen AI

Task: Propose your own unique problem/challenge and use the provided datasets to validate your solution. 

Context: We encourage students to solve complex, in-house problems where off-the-shelf solutions do not exist. Relevant examples of problems to identify can be within the dataset itself, overall healthcare industry, image quality, tracking success rate, disease prediction accuracy, etc.



Gen AI in healthcare


Generative AI models have the power to transform healthcare industry with its data-centric applications.

The technology has promising applications in drug discovery, vaccine development, and streamlining clinical operations.







Medical data (e.g. clinical notes, medical images) are a rich source of  information that can train generative AI models to understand diseases, patients, and treatments. 

By leveraging pre-training methods and combining text and image data, generative AI can create innovative solutions, including the generation of compound recipes for drugs, vaccines, and proteins. 


Project Background

C. Light will be providing you with the following types of datasets. Use this below information to help you form your processes to solve Track A, Track B, or Track X.


Data size: ~1M retinal images
Image resolution: 640 x 480 (Commercial) or 512 x 512 (Research)

  • 266/63/78 scans for MS/mTBI/MCI patients
  • Each scan contains 300 retinal images, 10 seconds video sequence with 122k retinal images in total
  • Each video sequence is paired with 2 x CSV files
  • 4 x datasets hosted on AWS S3 buckets, 3 datasets are disease specific, 1 dataset dedicated for testing image de-noising


Spatial resolution: 10 um
Motion tracking success rate: 90+%

  • Each video sequence has two associated CSV metadata: eye motion trace and statistics
  • Eye motion metadata gives and X (horizontal) and Y (vertical) relative movement in units of pixels and degrees (angular movement)
  • Statistics metadata include the metrics such as signal-noise-ratio, saccade (motion) amplitude, velocity, frequency, etc.


Disease states: Multiple sclerosis (MS), mTBI, mild cognitive impairment (MCI)
Disease prediction: 92% AUC

  • 46/32/7 Multiple Sclerosis (MS) / mild traumatic brain injury (mTBI) / mild cognitive impairment (MCI) patients
  • The EDSS provides a total score on a scale that ranges from 0 to 10 for MS disease.
  • Data for MS patients with EDSS score larger than 3.5 are provided in the dataset

Disclaimer: The above data is fake sample data and created specifically for the sole purpose of the 2023 Datathon and should not be treated as real research or commercial data. This data does not reflect the addressed disease groups but was created to only represent such in name only. Use outside of the intended event of Datathon is prohibited. Upon completion of the event, please destroy the data. If you have any further questions or concerns, please contact us at engineering@clighttechnologies.com with the subject line “Datathon help: [Add concise subject]”.

Download project datasets.

Download the track-specific dataset(s) to solve one of the two proposed problems (Track A or B) OR to solve a unique problem you originally defined (Track X).

It's your world and we're just living in it.

Pick the dataset(s) that makes most sense for the unique problem you have defined and plan to solve.

Download patient dataset 1 - MS (~20 GB)

Download patient dataset 2 - mTBI (~2 GB)

Download patient dataset 3 - MCI (~6 GB)

Download denoising dataset (~0.5 GB)

How we evaluate proposed solutions.

At C. Light, we value innovation and creative thinking.

We comprehensively evaluate projects based on criteria such as code efficiency, code style, and clarity of overall presentation – holistically assessing your thought process and conclusions.

We encourage you to challenge yourself by formulating and defining a unique problem to solve in lieu of the two problems initially proposed.

Whether you decide to address one of the two challenges we proposed or chart your own course to solve a challenge you craft yourself, our journey as AI scientists culminates in the fusion of innovation and precision.

The significance of algorithmic value underpins the indispensable role of artificial intelligence.











Evaluation metrics
to focus on:

  • Retinal image/video quality
  • Methods for enhancing model performance
  • Innovative approaches to validate embedded vectors & predict disease 


Eye motion and retinal video sequence are highly correlated, we can use this correlation to build a quantitative metric to evaluate model performance.​


Eye motion and retinal video sequence are highly correlated, we can use this correlation to build a quantitative metric to evaluate model performance.​


Both quantitative and qualitative measure of how the de-noising algorithm performs. For example, how do you define an image has a better quality than others?


Both quantitative and qualitative measure of how the de-noising algorithm performs. For example, how do you define an image has a better quality than others?


How do you measure if the "encoding" is working, or is the model really learning? Think about using a embedding vector from a trained model to generate synthetic retinal video sequence with associated eye motions as validation approach.


How do you measure if the "encoding" is working or is the model really learning? Can you use embedding vector from a trained model to generate synthetic retinal video sequence, with associated eye motions as validation approach?

FAQs by track.

We're collecting all relevant questions we get asked here to keep the playing field fair. 

Last updated 11/11/23 4:00pm PDT

What are saccades?

Small, involuntary jerk-like movements that occur during fixation

What is velocity?

Speed in a given direction. For the data, you can assume "velocity" as "average velocity".

What is peak velocity?

Maximum instantaneous velocity reached 

What is SNR?

SNR is signal-to-noise ratio. It refers to the relative magnitude of the signal compared to the uncertainty in that signal on a per-pixel basis.

Are good and bad images paired?

Unfortunately (but purposefully), the dataset provided is unpaired.  

Are there any references you recommend us to read/watch to get a better understanding of GANs, VAEs, and Diffusion Models?

Just a simple google search will be sufficient, but we have shared these papers: 1 , 2 

Does Track A mean more of building an interpretable classifier for good and bad images, or is it more of developing a denoising approach?

More denoising approach.

What are the images for track one? Like what do they represent?

Not relevant. Big hint: it is more of identifying how you'd objectively determine good from bad image and bad from goods images.

What causes noise in the bad images?

Some examples that can cause noise that produce bad images include, but aren't limited to:
  • optical artifacts/back reflection of eye
  • blink rate of patient
  • cataracts or related retinal diseases
  • sensor saturation

What’s the desired output of Track B?

The desired output for Track B is to generate synthetic retinal videos with associated, specific eye motions. 

What does "_raw" mean in the csv data for Track B?

For track B, focus on "_motion[deg]" and do not use "_raw[pixel]" and ignore “_raw”. “_raw” is most helpful for Track X.

No questions asked yet.


Strength in unity, change through community.


C. Light Technologies is a forward-thinking healthtech company whose mission is to innovate eye-tracking solutions using AI for brain and eye health and performance.

Active applications of retinal eye-tracking today includes medical diagnosis and treatment, drug development, human-computer interface (HCI), driver monitoring system, sports performance, and virtual and augmented reality (VR/AR).

science-backed innovation for healthcare.

Our Product


All your accurate, objective data to aid diagnosis

Our Product


10-sec Retitrack video scan of your eye.

Andrew Norton, PhD

VP, Optical Engineering

Andrew Norton, PhD is an optical engineer whose expertise lies in electrical engineering for adaptive optics and electro-optical technology development. He has a unique interest in wave optics and transforming the field of optics for use in neurologic spaces like MS, brain injury & TBI, and Alzheimer’s.
Andrew’s career has focused on developing state-of-the-art imaging systems for the visible, infrared & radar spectral bands.
Previously, he led the development of an experimental physics lab at Stanford University building high-resolution imaging systems & algorithms for ground-based astronomy telescopes & deep-tissue biological imaging applications.
Fun fact: Andrew has also designed a deployable large-aperture synthetic radar telescope for a stealth startup company & was the acting director of the Center for Adaptive Optics for 3 years!

Jacqueline Theis, OD

Chief Medical Officer

Jacqueline Theis, OD, FAAO is an award-winning optometrist with residency training in neuro-optometry and oculomotor dysfunction from UC Berkeley. 
Dr. Theis is an internationally-recognized speaker, published author, and key-opinion leader in her field for the diagnosis & management of vision problems and oculomotor dysfunction in concussion, acquired brain injury, & neurodegenerative disorders (i.e. multiple sclerosis, myasthenia graves, Parkinson’s and Alzheimers disease).
Her mission is to bridge the care gap between primary care, neurology and optometry/ophthalmology, and revolutionize the field(s) with active clinical research on the oculomotor system as a biomarker for neurologic function.
She was the Asst. Clinical Professor and founding Chief of the UC Berkeley Sports Vision and Concussion Clinic & currently practices at Virginia Neuro-Optometry within a transdisciplinary brain injury clinic.  She is an active member of IES, AOA, and AAO.

Christy Sheehy, PhD

Chief Executive Officer

Dr. Christy Sheehy co-founded C. Light Technologies & is the inventor of the FDA cleared (for marketing) product, Retitrack™.
She is an award-winning entrepreneur and healthcare innovator and published vision scientist with over 15 years of technical entrepreneurial expertise and direct clinical research experience with neurodegenerative patient populations at the top research institutions of UC Berkeley and UCSF. 
Dr. Sheehy successfully acted as the Principal Investigator for three NIH SBIR/STTR grants to study fixational eye motion changes in concussion and multiple sclerosis, and has brought in >$8M from venture capital, angel, and grant funding to the company to-date.
Dr. Sheehy’s mission is to create digital solutions using the retina to inform on brain health and improve prognostic care in all aspects of healthcare and sports vision performance.

Lon Dowell

Chief Commercial Officer

Lon Dowell is a dynamic sales and commercial executive leader who has cultivated 20+ years experience at the crossroads of healthcare, technology, and data.
With a strong background in the ophthalmologic sector, Lon’s expertise spans the domains of medical imaging, clinical workflow, and modality integration, establishing him as a distinguished authority and a sought-after speaker. His proficiency extends beyond individual contributions, as he excels in building & leading teams, propelling them towards the company’s overarching vision and the successful execution of business objectives.
Lon reflects the capability to navigate and thrive in a rapidly evolving industry landscape. He practices Business Philosophy that is focused on implementing a culture of responsibility: applying course corrections based on data & upfront success criteria models.

Joe Xing, PhD

Chief Technology Officer

Dr. Joe Xing co-founded C. Light and  20 years of experience in data science, machine learning, and artificial intelligence for smart optical design and related technologies.
Joe emerges as the strategic force to spearhead C. Light’s future growth in technology. His interests lies in leveraging cutting-edge technologies — such as artificial intelligence, machine learning, and data analytics — to drive C. Light’s innovation engine.
He leads the development and implementation of C. Light’s technology to academic and research institutions, clinics, and other medical offices. 
Joe is a part of Stanford University’s Y-Combinator and is an avid cyclist and golfer who enjoys life on the gravel and on the green.

Tracy Tran

Director, Investor Relations & Marketing

Tracy Tran has a diverse background spanning from clinical research management and operations for top research and academic institutions to life sciences and healthcare consulting for big pharma.
With a deep-rooted passion for healthcare technology and a multi-faceted perspective of the healthcare industry, Tracy recognizes C. Light’s immense potential to transform (primary) patient care.
Prior to joining C. Light, Tracy clinically trained at UC San Francisco and managed local and international clinical trials in retinal and neurodegenerative diseases for the Department of Neurology. Her previous clinical work aimed to address the pressing unmet needs felt in conditions where early symptoms often go undetected. She received her BA in Molecular & Cellular Biology from UC Berkeley.
For C. Light, Tracy takes on tasks related to investor relations & strategic partnerships, marketing, PR, and clinical research mgmt.

Our Product


Physicians neuro toolkit can’t capture it .