machine learning – Computational Ophthalmology

July 1, 2021September 1, 2021

Student becomes teacher: training faster deep learning lightweight networks for automated identification of optical coherence tomography B-scans of interest using a student-teacher framework

In this study, Dr. Julia Owen and her co-authors developed a novel method for automated detection of abnormal optical coherence tomography (OCT) B-scan ("B-scan of interest") images of the retina. Commercial OCT devices do not routinely provide automated diagnoses, so disease detection requires expert interpretation of the images by an ophthalmologist. Reliable automated detection of potentially abnormal retinal scans could save time, as the machine could flag abnormal scans that require further review by a clinician. But there have been challenges with developing such models, including the lack of large training datasets, standardized methods for image acquisition and processing, agreed-upon evaluation metrics, and limitations in computing power.

Dr. Owne developed an approach that overcomes some of these limitations, by using an efficient method for training lightweight machine learning models, or models with fewer parameters and operations, using unlabeled and routinely-available data. Lightweight models are useful in clinical settings because they can perform quickly and are more compatible with portable devices.

July 11, 2020October 7, 2020

Protecting Data Privacy in the Age of AI-Enabled Ophthalmology

Journal

Citation

In this commentary in the journal Translational Vision Science and Technology, Dr. Elysse Tom and her co-authors discuss digital data privacy in the era of "Big Data" and artificial intelligence. Although these new technologies offer many potential benefits for patients and for healthcare systems, there is a need for better data protection methods that can evolve with these advances.

The authors discuss some of the ethical issues associated with the use of these increasingly large clinical datasets, noting that maintaining data privacy and confidentiality-and thus respect for persons-is a challenge. For example, breaches of patient privacy can cause major harms and can also have unintended consequences, such as impacting employment or insurance coverage or allowing hackers to obtain Social Security numbers and personal financial information.

June 1, 2020October 9, 2020

Clinical Applications of Continual Learning Machine Learning

Journal

Citation

In a recent commentary in The Lancet Digital Health, Dr. Cecilia Lee and Dr. Aaron Lee address artificial intelligence and its role in the practice of medicine, specifically potential uses for continual learning models. Continual learning is a type of machine learning in which the model continues to learn and improve as it receives new data while retaining what it has previously learned.

This ability to continuously fine-tune their performance makes continual learning models better suited for certain complex tasks than the more traditional machine learning devices used in clinical practice today. But the fact that they constantly update and change also makes these models more complicated to implement into patient care systems. The authors discuss the practical, ethical, and regulatory issues that must be addressed before these models can be fully integrated into clinical practice.

February 15, 2020October 15, 2020

How Artificial Intelligence Can Transform Randomized Controlled Trials

Journal

Citation

In this commentary for the journal Translation Vision Science and Technology, Dr. Aaron Lee and Dr. Cecilia Lee discuss how the application of artificial intelligence and big data in healthcare has the potential to transform clinical trial research.

They explain some of the common pitfalls of randomized controlled trials and how deep learning is particularly well-suited to overcome some of these problems, leading to more efficient execution and greater statistical power than what would be expected from traditional trials.

February 5, 2020July 9, 2020

Logistic Regression Classification of Primary Vitreoretinal Lymphoma Versus Uveitis by Interleukin 6 and Interleukin 10 Levels

Pubmed

Journal

Citation

In this study published in the journal Ophthalmology, the authors evaluated the use of a logistic regression model for early diagnosis of primary vitreoretinal lymphoma. Primary vitreoretinal lymphoma is a rare disease with a generally poor prognosis. Early diagnosis of local ocular disease has been shown to prolong survival significantly, but because the disease is rare and the ocular symptoms are nonspecific, it is often misdiagnosed. In addition, although cytologic analysis of aqueous or vitreous samples is diagnostic, there are often problems with lymphoma cell detection in the sample. As a result, the diagnosis of PVRL is often delayed, taking on average 1 to 2 years from the onset of symptoms and typically requiring multiple biopsies.