Amazec recycles ASML technology to diagnose heart failure

Technology originally developed to measure vibrations in ASML equipment can also be used to diagnose heart failure. It got Technobis founder Pim Kat so excited that he’s postponing retirement to bring it to the market.

It seemed like a really good idea, but Pim Kat wanted to be sure. Let’s find out whether doctors would really have a need for our medical device, he suggested to his potential business partners. And so they sent out questionnaires to one hundred experts around the world, soliciting their opinion on the concept.

Kat set the bar high: if he was going to shore up his retirement, he wanted a response rate of at least 70 percent, of which 90 percent needed to be positive. He got a 130 percent response rate – many doctors enthusiastically forwarded the questionnaire to colleagues – of which 97 percent was overwhelmingly positive. No retirement yet for Kat, who turns 65 in a few weeks.

He was well on this way, though. Last year, Kat sold a majority stake in instrumentation specialist Technobis, the company he co-founded in 1996, and relinquished his responsibilities there. More recently, he had also been dialing down his duties as CTO at Photonfirst, the photonic-sensing company that was carved out of Technobis at the start of this year and in which he retains a stake as well.

But now Kat has started a new company after all, Amazec Photonics, to market an instrument that will greatly improve the ability to measure how well the heart functions. This is particularly useful for the diagnosis of heart failure, a condition in which the heart muscle doesn’t pump blood as well as it should. In addition, the device will enable doctors to determine the amount of blood in circulation within a few minutes. This is extremely valuable information, for example, when someone suffered heavy blood loss after an accident.

“I just didn’t have the heart to pass up on this beautiful application,” quips Kat. Moreover, he doesn’t expect that he’ll be tied up in his new business for years: the technology has already been developed for another application – by Technobis, in fact – and the proof of principle has been demonstrated. With a target of introducing the first devices in the field in early 2023, there will still be plenty of retirement left for the Amsterdam-born entrepreneur.

Blood volume

The story of Amazec starts with a conversation a couple of years ago between Ton Backx, dean of the Electrical Engineering department at Eindhoven University of Technology (TUE), and Erik Korsten, an anesthesiologist-intensivist at the Catharina Hospital in Eindhoven. Korsten told Backx how he routinely used to inject a 4 °C physiological saline solution into the veins of suspected heart failure patients. The solution travels to the heart, and the way it behaves in there is a measure of heart function. This is registered through a catheter fitted with a temperature probe, which is inserted into the pulmonary artery.

The more efficient the heart, the quicker the ‘cold blob’ will pass through it. As a consequence, the blob will stay reasonably intact when a heart is strong. With a heart that does a lot of vibrating instead of pumping, on the other hand, the blob will ‘dissolve’ into a larger volume of warm blood. This dilution process can be measured by the temperature probe, yielding so-called dilution curves.

A typical dilution curve measured this way doesn’t give very detailed information, however. The registered dip in temperature is only 0.8 °C, and the probe has a resolution of 0.1 °C. “Basically, traditional temperature-based dilution measurements show whether the heart is functioning well, not well at all, or something in between. That’s about it.” For this reason, and the fact that even minimally evasive procedures aren’t without risk, the method has fallen into disuse.

Backx suggested that Korsten contact Kat at Technobis, which he knew had developed high-resolution temperature sensing technology (see inset “Integrated photonic temperature sensing”). Korsten was reluctant, but after some gentle nudging by Backx, he begrudgingly made the trip to Alkmaar. “He told me that he agreed to meet me because he didn’t want to be impolite,” Kat laughs.

Kat took his guest to see the setup that Technobis, commissioned by ASML, built to measure nanovibrations in semiconductor manufacturing equipment. However, the technology is equally capable of measuring temperature – if so desired, up to a millionth of a degree Celsius. “I showed how simply moving my hand up and down near the sensor probe produces a wave-like output. That’s caused by the heat radiating from my hand.”

Korsten was sold immediately. Eager to further explore the possibilities, he built a circulation system using parts of a heart-lung machine and tubing. Together, Kat, Korsten and his Catharina Hospital colleague Arthur Bouwman tested whether they could detect an injected cold blob passing the pump. That proved no problem at all. In fact, due to the ultrahigh sensitivity of the sensor, they could see the blob passing it up to seven times.

“Next, we tried it on Korsten. Bouwman injected him with the cold saline solution. We couldn’t insert the probe into the lungs, so we simply put the photonic sensor on the skin overlying an artery on the wrist. Well, that worked wonderfully too! We could see a cold blob injected into one hand passing through the wrist of the other arm, with a resolution of a ten-thousandth of a degree Celsius. So we could obtain much more accurate dilution curves, without having to insert a probe into the body – which is obviously uncomfortable for the patient.”

Thanks to the much higher resolution, the experimenters picked up on all kinds of unexpected things. As it turns out, 10 millimeters of cold liquid doesn’t equilibrate with its surroundings as quickly as one would expect: the blob can be seen passing the probe up to four times. In addition, respiration and even the heartbeat show up in the ultrasensitive blood temperature measurements. “If someone takes a breath of cold air, from a freezer for example, you can see it passing the probe up to three times. So perhaps even the injection of cold liquid may ultimately be unnecessary.” All these findings were backed up by computer-modeling experiments performed by Backx.

Finally, Kat and partners realized that their technology could also be used for other applications. Using the dilution curve integrals of the blob passing multiple times, for example, it’s quite easy to calculate the blood volume – a piece of information that doctors craved to have easy access to for decades. There’s probably a lot more happening in the human body that the photonic temperature probe can uncover, Kat suspects.

Clinical validation

After the proof-of-principle experiments and the overwhelmingly positive questionnaire response, Kat decided to start Amazec, together with Korsten, Backx and Bouwman. Appealing to some friends, Kat closed the first funding round within two weeks. Another round will be needed to develop fully functional prototypes for field testing.

“Most of the work ahead is medical in nature: clinically validating our device,” explains Kat. “Before we can start testing on humans, we’ll have to perform lab experiments on isolated hearts. And we’ll probably do animal testing as well.” The technology itself is basically available off the shelf – Kat will become a customer of the two companies he founded. “The photonic integrated circuits have been developed for the semicon project. We can order them from Photonfirst. Technobis will build the instrument around it. It has extensive experience in the medical field, so certification won’t be a problem.”

“Getting to a marketable product will be relatively low-effort. I’ll do the project management and the doctors will oversee the clinical validation. We won’t need any staff or anything like that. Once we’re nearing market introduction, our target is to be acquired by a major medical company. It’s not entirely impossible, however, that we’ll build Amazec into a full-scale business,” Kat says. Perhaps no retirement after all?