Professor Anders Heijl: The International Perimetric Society met in Sacramento in October of 1982. During that meeting, we learned there that Humphrey Instruments, a US company later acquired by ZEISS, had engineered and built breadboard prototypes of an automated perimeter. This was a meeting that almost everyone in the world who was working on computerized visual field testing attended. Humphrey sent several employees to the IPS meeting and they talked to those of us who had experience in that field. I had nine years of experience at that time and I had already built two perimeters, but I was interested in implementing more of my ideas. Humphrey found it worthwhile to collaborate with me. I flew home to Sweden, and Mike Patella and I started corresponding and working together very soon thereafter. We have continued that collaboration ever since.
Dr. Mike Patella: Before we approached Professor Heijl, I reviewed almost all of the medical literature on visual field testing. That review clearly showed that collaboration with Professor Heijl was essential, and time has proven us right. Without Anders, nothing would have worked out the way it did.
Patella: The HFA was first presented as a demonstration breadboard at the 1982 meeting of the American Academy of Ophthalmology in San Francisco. The following year, we delivered twelve clinical-grade engineering prototypes, one to Professor Heijl and the rest to key eye doctors in North America. I collaborated closely with my ZEISS-Humphrey colleague, Christine Ritter, who was the very first HFA product manager, and she and I worked intensely with the Swedish team led by Professor Heijl and Professor Boel Bengtsson at the University of Lund.
Dr. Mike Patella
Before we approached Professor Heijl, I reviewed almost all of the medical literature on visual field testing. That review clearly showed that collaboration with Professor Heijl was essential, and time has proved us right.
Heijl: Perimetry is the second most-common visual function test that eye doctors use, after visual acuity measurement. It is most often used to diagnose and treat patients with glaucoma, but also in neurological diseases and retinal disorders.
Patella: The success of the HFA had little to do with the hardware design – that was always the simplest part of it. Success was a result of at least three factors: collaboration with leading eye doctors worldwide, development of maximally efficient testing algorithms, and invention of novel data analysis methods.
Heijl: Perimetry measures how well a patient sees at various locations in his or her peripheral vision. Effective perimeters quickly and accurately find the limits of what the patient can see in a set of predetermined visual field test points. These findings are compared to typical ranges of normal vision throughout the visual field, and areas with sub-normal sensitivity are identified as visual field defects. The shape and location of these defects can vary depending on the disease and disease stage. Often, measuring the visual field and quantifying these defects can give the doctor the diagnosis even before the rest of the eye has been examined.
Patella: When first launched in 1984, HFA tests took 15 minutes per eye, and results were difficult to interpret, since they were just raw visual sensitivity numbers. Today, tests can be completed in as little as two minutes per eye and test reports include automatic plain language analyses telling the user whether each test fell within or outside of normal limits and whether or not series of baseline and follow-up tests show statistically significant change over time.
Much of this work was unprecedented in ophthalmology. For instance, one of our early projects was to collect visual field tests of normal people at multiple sites in the US and also in Sweden. As far as I know, that project was ophthalmology’s first large scale international collection of normative data. We called the resulting analysis package Statpac. Today, we might refer to parts of Statpac as Artificial Intelligence, but thirty years ago we weren’t thinking about our work in those terms. Successfully getting testing times down to two minutes without losing accuracy has been a thirty-year journey, as you might have guessed.
Heijl: When we first launched the HFA, we understood perfectly well that our very next step had to be to simplify and standardize interpretation of test results, and that understanding led to the release of Statpac, in 1986. This was our first comprehensive analysis package, and it introduced several concepts that have become worldwide standards. One was the so-called probability maps, which facilitate differentiating between true field loss and typical measurement variability.
A second new concept was pattern deviation, which helped reduce or eliminate the effects of cataract on test results. Interpretation was further standardized and simplified with the introduction of the Glaucoma Hemifield Test – or GHT – a few years later. The GHT produced a plain language analysis of whether or not the overall test was within normal limits. A few years after that, we developed Change Probability Maps, which today are used worldwide to detect statistically significant worsening of the visual field over time. Change Probability Maps were used as a clinical endpoint for the landmark Early Manifest Glaucoma Trial (1994–2013).
Around 2006 we developed the Visual Field Index, which we believe to be one of the best methods for measuring the rate of visual field progression in glaucoma, and today is one of the most important elements in deciding on the individual treatment needs of glaucoma patients.
Test efficiency took a huge step forward with the release the SITA (Swedish Interactive Threshold Algorithm) family of test strategies in 1996. Development had taken ten years, and the algorithms were so computationally-intensive that they could not be used with the computers that were available when development first began around 1987. But we all knew that computers would soon be much faster and much less expensive, and by the time that SITA was ready for release the computers we needed were indeed available and affordable. Development did not stop there. A newer and faster version of SITA was released just one year ago.
Heijl: Soon after it was released, HFA became the world’s most widely used perimeter. It has taken more than two thirds of the world’s market ever since and has become a gold-standard instrument. Patella: If you have glaucoma anywhere in the world, your disease probably will be managed using an HFA. Of course, your doctor will also consider information from other instruments, as well as his or her clinical examination, but HFA probably will play an important role in your care. More generally, the existence of a single widely used perimeter has allowed doctors to speak a common language worldwide when discussing the effects of blinding eye diseases on visual function.
Patella: If you have glaucoma anywhere in the world, your disease probably will be managed using an HFA. Of course, your doctor will also consider information from other instruments, as well as his or her clinical examination, but HFA probably will play an important role in your care. More generally, the existence of a single widely used perimeter has allowed doctors to speak a common language worldwide when discussing the effects of blinding eye diseases on visual function.
Prof. Anders Harald Robert Heijl
Soon after it was released, HFA became the world’s most widely used perimeter. It has taken more than two thirds of the world’s market ever since and has become a gold-standard instrument.
Patella: The HFA is best thought of as a work in progress, and a work that is more about applications software than about hardware. We do update the hardware from time to time, but this is mostly done when new software outgrows current hardware capabilities. Today, after 35 years of continuous refinement, there still is much more that can and should be done. Anders & I and our partner, Professor Boel Bengtsson – and our many collaborators – are not at all finished. This story is by no means over.
Heijl: For example, we recently released SITA Faster, a new test strategy that was given at no charge to all HFA3 owners. And we can think of many ways to provide even better interpretation programs. Only three versions of HFA hardware have been released over the 35 years we have been in the market, demonstrating that the hardware has not been changed very often. But there have been frequent and regular software updates. If the HFA continues to be as successful as it is now, there will of course be an HFA4 at some stage. ZEISS is the perimetry market leader, and that comes with the responsibility for continuing research and development.
Heijl: That’s probably true to some small extent, but only for eye care professionals and their patients.
Patella: That is what we and the many colleagues who joined us along the way were hoping to do. If we have succeeded, it is only because we had lots of help.
Dr. Mike Patella began his career as an aerospace engineer, working on NASA manned spaceflight projects. He left aerospace in order to become an optometrist and concurrently joined a Berkeley startup called Humphrey Instruments, which had been founded by Nobel Laureate Luis W. Alvarez. Humphrey was later purchased by Zeiss. Dr. Patella worked for more than four decades on the development of automated ophthalmic diagnostic instruments, including refractive instruments, contact lens fitting devices, imagers, and perimeters. Mike retired from Zeiss in 2018.
Today Mike lives in Alamo California. He is a senior partner at Sand Hill Consulting Associates in Santa Rosa, CA, a perimetry consultant to Zeiss, and a Visiting Scientist at UC Berkeley’s School of Optometry. He is a member of the American Glaucoma Society, a fellow of the American Academy of Optometry, and co-founder of the Optometric Glaucoma Society.
Professor Anders Harald Robert Heijl is senior professor and consultant at the department of Ophthalmology at the Skåne University Hospital in Malmö and the University of Lund. He started working on computerizing perimetry – his PhD project – under the leadership of professor Torsten Krakau in 1973, and he and his research group have continued working in that field ever since – in recent years under the leadership of professor Boel Bengtsson. Some of those perimetric developments were essential for important prospective glaucoma clinical trials, and Heijl was the Study Director and Primary Investigator for the landmark Early Manifest Glaucoma Trial (EMGT) study. Heijl is an internationally recognized glaucoma expert, has been President of the International Perimetric Society and the Glaucoma Research Society, has served as the ophthalmic advisor to the Swedish Board of Health and Welfare, has received more than 25 scientific awards, and has been made an honorary member of four ophthalmological societies.