Endophthalmitis Rate in Immediately Sequential versus Delayed Sequential Bilateral Cataract Surgery within the Intelligent Research in Sight (IRIS) Registry Data

This large retrospective cohort study used data from more than 5 million people who underwent bilateral cataract surgery in the United States to investigate the rate of post-operative endophthalmitis. The authors compared the rates of endophthalmitis in the group of patients who had both eye surgeries on the same day (immediate sequential bilateral cataract surgery, or ISBCS) to the group of patients who either had cataract surgery in both eyes on separate days (delayed sequential bilateral cataract surgery) or who just had surgery in one eye. The authors found that after controlling for age, sex, race, insurance status, and comorbid eye diseases, the risk of postoperative endophthalmitis was not statistically significantly different between these two groups, . 

Continue reading "Endophthalmitis Rate in Immediately Sequential versus Delayed Sequential Bilateral Cataract Surgery within the Intelligent Research in Sight (IRIS) Registry Data"

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.

Continue reading "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"

Refractive Outcomes After Immediate Sequential vs Delayed Sequential Bilateral Cataract Surgery

In this large population-based study using data from  American Academy of Ophthalmology Intelligent Research in Sight (IRIS) Registry, Dr. Cecilia Lee and her co-authors looked at visual acuity outcomes after bilateral cataract surgery. Specifically, they compared outcomes after immediate sequential bilateral cataract surgery or ISBCS (patients undergo surgery for both eyes on the same day, as separate procedures), and delayed sequential cataract surgery or DSBCS (patients have the surgeries on separate days). ISBCS is less commonly performed in the United States, but is growing in popularity and may have some advantages such as fewer follow-up visits, immediate vision correction, and lower cost. However, many ophthalmologists prefer delaying the second surgery in order to assess vision after the first surgery, in order to make adjustments for the second eye.

Continue reading "Refractive Outcomes After Immediate Sequential vs Delayed Sequential Bilateral Cataract Surgery"

Application of deep learning to understand resilience to Alzheimer's disease pathology

The term "resilient" is used to describe the unique set of people who develop the neuropathological features of Alzheimer's disease but do not show signs of the cognitive decline that is typically associated with the disease. In this study, published in the journal Brain Pathology, Dr. Cecilia Lee and Dr. Aaron Lee used a machine learning approach to identify subtle differences in brain tissue of "resilient" patients compared to patients with Alzheimer's disease. These patients had been participants in the Adult Changes in Thought study and donated their brains for research into the causes of Alzheimer's disease and related dementias.

Continue reading "Application of deep learning to understand resilience to Alzheimer's disease pathology"

Recommendations for Standardization of Images in Ophthalmology

In this editorial, writing for the American Academy of Ophthalmology, Dr. Aaron Lee and his co-authors explain the urgent need for ophthalmic imaging device manufacturers to standardize their imaging formats to comply with existing international standards. Currently, manufacturers of devices such as optical coherence tomography machines use their own proprietary imaging formats, requiring special software to access and analyze the images obtained with their device. This common practice makes it difficult for clinicians and researchers to compare images from different machines. Standardization would allow interoperability between imaging systems, allowing electronic health information to be transferred more easily when a patient is seen at different hospitals or clinics. It would also allow researchers to build comprehensive imaged datasets for and big data analyses and machine learning studies, a growing area of research in ophthalmology that is currently limited by lack of image standardization.

Continue reading "Recommendations for Standardization of Images in Ophthalmology"