Article

Manage Images and Data in Practice

What O.D.s need to know in the evaluation of imaging management systems

In today’s health care environment, the use of integrated, multimodal image and data management systems can set the stage for better patient outcomes. Consider how both optometrists and ophthalmologists have greatly expanded their use of integrated multimodal diagnostic imaging. OCT and VF progression analysis demonstrate that both structural and functional data should be closely monitored across the entire expanse of ocular disease severity (glaucoma, diabetic retinopathy [DR] and AMD).1,2 Integration with en face OCTA detection of microvascular flow abnormalities and invisible subclinical tumors exhibit the value of multimodal imaging in aiding in the identification of causes of visual loss.3

However, without a method by which to integrate, analyze and store these images, the optometrist may be at a disadvantage to fully realize the benefits of this dynamic technology. Here, I’ll discuss what optometrists need to know to evaluate medical imaging systems.4

ACCESS TO A FULL RANGE OF FEATURES

The advent of digital imaging allows for rapid access of ophthalmic images, including comparison of multiple images. To take full advantage, an image management system should be able to utilize and target dynamic visualization of raster scans, cubes, medical images (photos, videos, angiography), heat maps and multiple exam data.

It is important that any image and data management system can easily access electronic health records (EHR). O.D.s should check to ensure the system they are evaluating will integrate with the EHR system they use, so as to easily manage and store images within the medical record.

ASSIST IN ANALYZING IMAGES

Rapid advances in OCT, OCTA, swept-source OCTA (SS-OCT-A) and the fractal dimension of an OCTA image have armed the modern optometrist with new powerful diagnostic tools to examine micro-vasculature, including deep choroidal vasculature.5,6 The ability to enhance images (increasing contrast, brightness, resolution) and compare them with fundus autofluorescence, among other technologies, is an important characteristic for imaging and data management systems in the diagnosis and management of retinal disease. Integrated multimodal imaging systems have the powerful ability to allow viewing of images side-by-side, dynamic multilayer viewing, and they allow for multiple examination comparisons.

Additional analysis functions include:

Progression analysis. The use of structure-function progression analysis software in the management of glaucoma is well established and also is being applied to DR and AMD.2,7,8 Structure-function summary reports will help the clinical optometrist in both the diagnosis and management of ocular disease.

AI-enabled algorithms. The ability of imaging management systems to monitor ocular disease progression and to recognize important ocular biomarkers for future AI-enabled algorithms should be a core feature of any system.9 An example of this technology at work is the grading and automating of DR.

Automated DR assessment. The ability of integrated imaging management systems to allow for accurate grading of DR is an important feature.10 Automated DR image assessment software has shown acceptable sensitivity for referable DR when compared to human graders and manual grading.11 The use of retinal photography and human grading in DR screening can be very labor intensive and expensive (hardware/software, training, etc.). Automated retinal image analysis software may offer an alternative to manual human grading.10

freshidea/stock.adobe.com

Better Patient Outcomes

The global COVID-19 crisis has exposed all vulnerabilities in modern society, including those associated with the practice of optometry. How does optometry mobilize to care for our diverse communities and decrease the many worldwide health care disparities?

“The COVID crisis has shown us all that our care sites do not have good patient data, do not have good patient linkages, usually do not have team care of any kind in place and most are so dependent on current piecework fee volumes from patients, that they quickly collapse financially when that volume is interrupted,” former Kaiser Permanente CEO George Halvorson states in The Health Care Blog.

Eric Topol, M.D., predicts the democratization of medicine in his book, The Patient Will See You Now: The Future of Medicine Is in Your Hands, which sees a convergence of science, technology and society.

As we continue into the 21st Century, the year 2020 and the global COVID-19 pandemic, new innovative technologies, such as the ones mentioned in this article, can set the stage for better patient outcomes.

PRIVACY, SECURITY AND OPERATING STANDARDS COMPLIANCE

HIPAA compliance and the ability to maintain patient privacy is a must in modern optometric care. Modern trends in telehealth and patient-driven care make image sharing among an optometrist, other clinicians and patients an important feature for any image management system. Furthermore, interprofessional collaborations make the importance of removing protected health information from images an important feature.

Compliance with the Digital Imaging and Communications in Medicine (DICOM) standard (dicomstandard.org ) is an important feature for all modern imaging and data management systems. O.D.s should ask for an imaging system’s DICOM conformance statement (study DICOM at aao.org/mit ). When discussing imaging systems with vendors, some ophthalmic devices and systems are “DICOM capable,” but are not bidirectional in their interfaces.12

Systems able to use multiple devices from a variety of companies using the Picture Archiving Communication Standard (PACS) to be vendor neutral are also helpful and can reduce our need to purchase and maintain duplicate devices. The inability to easily store and access our images will greatly reduce any system’s value to our practice. Time taken to examine and design our imaging workflow will be time well spent, as it will improve our time efficiency and move our practice toward more value-based optometric care.

OPHTHALMIC IMAGE ANNOTATION

Modern optometrists will find the ability to annotate findings in an integrated imaging data management system useful in their busy workflows. Annotation will support the diagnostic and procedural billing codes and improve interprofessional communication and patient education and reduce the dependence on viewing original images on our ophthalmic devices.13,14 Furthermore, they will set the stage for future AI deep-learning algorithms.15

STORAGE SPACE AND NETWORK INFRASTRUCTURE

As a brief, yet related, aside, investing in storage space and network infrastructure is a must. New technological advances offering high-definition photos and videos will create an increasing need for storage and network infrastructure. Our hardware capacity needs will grow as we advance our imaging workflows and add more devices. Let’s be prepared to interface smartphones, tablets, wearables and other devices to our image management systems.

LET’S DO OUR HOMEWORK

When evaluating integrated image management systems, let’s do our homework; examine our current workflow, including our ophthalmic devices/networking infrastructure and account for the industry standards HIPAA, DICOM and PACS compliance. Integrated multimodal image and data management systems offer advanced outcome support in the management of ocular disease. Furthermore, they may offer methods of incorporating disease grading (e.g. glaucoma, DR, AMD) and both automated and autonomous AI algorithms in the future. OM

Available Integrated Imaging & Data Management Systems:

THERE IS A GROWING LIST of integrated ophthalmic imaging and data management systems available for optometrists today including:

ZEISS FORUM: Carl Zeiss Meditec Inc.

EYESUITE: Haag-Streit Diagnostics

SYNERGY: Topcon Medical Systems

HEYEX 2: Heidelberg Eye Explorer

NETVUE: Optovue

EYECLINIC: MaximEyes EHR

This list will be updated at optometricmanagement.com .

REFERENCES

  1. Bayer A. (2018) Combining structure and function in glaucoma. In: Akman A., Bayer A., Nouri-Mahdavi K. (eds) Optical Coherence Tomography in Glaucoma. Springer, Cham. doi.org/10.1007/978-3-319-94905-5_16 .
  2. Nguyen AT. Greenfield DS. Bhakta AS, Lee J, Feuer WJ. Detecting glaucoma progression using guided progression analysis with OCT and visual field assessment in eyes classified by International Classification of Disease Severity Codes. Ophthalmology. 2019 125;4:479-482. doi.org/10.1016/j.ogla.2018.11.004
  3. Mellen P, Sioufi K, Shields J, Shields C. Invisible, Honeycomb-like, Cavitary Retinal Astrocytic Hamartoma. Retinal Cases & Brief Reports. 14. 1. 10.1097/ICB.0000000000000697.
  4. Burling-Phillips L. Image management systems: what you need to know. EyeNet Magazine. June 2013. Accessed July 9. https://www.aao.org/eyenet/article/image-management-systems-what-you-need-to-know
  5. Schaal KB, Munk MR, et al. Vascular abnormalities in DR assessed with swept-source optical coherence tomography angiography widefield imaging. Retina. 2019;39(1):79-87.
  6. Bhardwaj S, Tsui E, Zahid S, et al. Value of fractal analysis of optical coherence tomography angiography in various stages of diabetic retinopathy. Retina. 2018;38(9):1816-1823.
  7. Norgaard MF, Grauslund J. Automated screening for diabetic retinopathy – A systemic review. Ophthalmic Res. 2018;60:9-17.
  8. Kanagasingam Y, Bhuiyan A, Abramoff MD, et al. Progress on retinal image analysis for age related macular degeneration. Prog. Retin. Eye Res. 2014;38; 20-42. doi.org/10.1016/j.preteyeres.2013.10.002
  9. Gargeya R, Leng T. Automated identification of diabetic retinopathy using deep learning. Ophthalmology. 2017;124(7):962-969. doi: 10.1016/j.ophtha.2017.02.008. Epub 2017 Mar 27.
  10. Olvera-Barrios A, Heeren TFC, Balaskas K, et al. Diagnostic accuracy of diabetic retinopathy grading by an artificial intelligence-enabled algorithm compared with a human standard for wide-field true-colour confocal scanning and standard digital retinal images. British J of Ophth. Published Online First: 06 May 2020. doi: 10.1136/bjophthalmol-2019-315394
  11. Tufail A, Rudisill C, Egan C, et al. Automated diabetic retinopathy image assessment software: diagnostic accuracy and cost-effectiveness compared with human graders. Ophthalmology. 2017 Mar;124(3):343-351. doi: 10.1016/j.ophtha.2016.11.014. Epub 2016 Dec 23.
  12. Smith EM, Ruffel JD, Fisher M. A generic digital imaging and communications in medicine solution for a bidirectional interface between the modality and the radiology information system. J Digit Imaging. 1999;12(2 Suppl 1):93-95. doi:10.1007/BF03168767
  13. Pilla L. The complete picture. Ophthalmology Management. 2019. 23;16-17,30.
  14. 5 must-have ophthalmology image management software features. Modernizing Medicine. Accessed July 9. https://www.modmed.com/blog/ophthalmic-image-management-must-haves-2/
  15. How to use semantic image segmentation annotation for medical imaging datasets? Medium. Published April 6. Accessed July 9. https://medium.com/anolytics/how-to-use-semantic-image-segmentation-annotation-for-medical-imaging-datasets-dba536b018df