The sweet, tart flavor of a ripe tomato is one of those pleasures of summer. Just not to Rob O'Reilly.
A senior engineer at Analog Devices Inc., O'Reilly has been trying to improve the taste of tomatoes, even though he can't stand them. "Had it been grapes, I'd have been much happier," said O'Reilly, who enjoys a glass of wine now and then.
Nonetheless O'Reilly is building a network of electronic sensors — an "Internet of Tomatoes" — to measure heat, light, and moisture levels for clues on how to improve flavor and boost yield.
He is also working with an Israeli company on a gadget that can judge the quality of a tomato simply by hitting it with a beam of light. At Analog Devices they call it a "tomato tricorder," like the hand-held scanners on the TV series "Star Trek."
Helping farmers grow a higher-quality tomato could substantially boost their revenue; most local tomatoes, O'Reilly sniffed, are of mediocre quality that sell for just pennies a pound.
"The project goal is absolutely about helping local farmers," O'Reilly said.
The Wilmington company specializes in microchips for cellphones, cars, and spaceships, and O'Reilly is an expert in digital sensors. But he stumbled into the tomato patch while attending a conference in Phoenix and heard Francis Gouillart, chief executive of Stock Pot Malden, an incubator that helps launch specialty food businesses, urge the tech industry to make digital devices to improve food quality.
"When I came down from the stage, Rob O'Reilly was there," Gouillart recalled. The two men decided that the tomato trade was particularly ripe for a technology upgrade. For one thing, there are plenty of tomato growers in Massachusetts. For another, it's hard to grow good tomatoes, so farmers would probably be eager to work with them.
Soon, O'Reilly and a fellow Analog Devices engineer, John Patrick O'Connor, were up to their elbows in tomatoes, some purchased at stores, others provided by local farmers such as Steve Verrill, co-owner of Verrill Farm in Concord. Though he raises "everything from asparagus to zucchini" on his 150 acres, Verrill said tomatoes are the main cash crop. He hopes O'Reilly's research will establish a scientific standard for tomato quality.
"It's a very subjective thing, and this is the first time I've seen a way to quantify it a little bit," Verrill said.
In his research, O'Reilly purees one tomato at a time and then uses an optical scanning system that calculates the amounts of sugars and acids in it. He brought in professional chefs to do something O'Reilly would never do: taste them. By comparing the chemical makeup of each tomato with the opinions of the tasters, O'Reilly and O'Connor assembled an objective database of tomato quality, in order to predict its flavor solely by knowing its chemical composition.
Last August, O'Reilly and O'Connor set up a booth at the annual Boston Tomato Contest, where they tested samples and said they successfully predicted the champions later picked by judges.
"We did a whole bunch of chemistry and optical measurements," O'Reilly said. "Between the two of us we actually picked the winner before anybody tasted them."
Farmers lack a portable laboratory that can analyze tomatoes in the field, but perhaps not for long. In February, Analog Devices formed an alliance with Consumer Physics Inc., an Israeli company that's built a hand-held version of the mass spectrometers used by chemists. The SCIO is a $250 device that bathes its target in light and measures the sugar, salt, and acid content of a tomato.
The SCIO reading "is essentially a fingerprint, which is then sent through a smartphone or nearby device to the cloud," said Consumer Physics chief executive Dror Sharon. An Internet-based data center analyzes the chemical fingerprint and displays the results on the user's smartphone. SCIO was designed to test all sorts of fruits and vegetables, as well as drug compounds.
O'Reilly has linked SCIO to his tomato chemistry database, so the fingerprint from a scanned tomato will accurately predict its flavor.
"With this unit, I don't have to smash the tomato anymore," O'Reilly said. "I can literally just jog up to it, fire it at the tomato, and it'll give me a list of parameters that are important to me."
It could do the same for a farm worker trying to decide which tomatoes are ready for harvest. And Analog is designing a version of SCIO small enough to fit into a cellphone, so that a consumer could run the same test while strolling through the produce department.
But to produce good tomatoes in the first place, O'Reilly wants to track their entire growing cycle. So he's set up a miniature vegetable patch on Analog Devices' Wilmington campus. Above each group of tomato plants hangs a beige box roughly the size of a pack of gum. Code-named "Fenway," the box contains sensors that measure temperature, humidity, and the brightness of the sun, and the transmitted data are displayed on a large TV screen in O'Reilly's nearby office.
"It tells you when to water, when not to water," his partner O'Connor said. "When to harvest and when not to harvest."
This level of detail can make a big difference in crop quality. For example, farmers use a simple formula called "growing degree days" to know when it's time to harvest a crop. Air temperature is one factor — higher temperatures usually mean an earlier harvest. Usually, farmers use temperature data provided by local weather reports. But on a recent sunny day at the Wilmington tomato patch, the Analog Devices sensor measured the temperature around the plants at 100 degrees, much warmer than the 83 degrees for the area reported by the National Weather Service.
The more precise data would allow farmers to pick tomatoes at the ideal moment.
Food producers are expected to spend more than $18 billion on "smart agriculture" systems by 2022, much of it for sensors that tell farmers exactly what's happening to their crops, according to research firm Markets and Markets.
So if O'Reilly's technology is successful, that could open a huge new market for Analog Devices.
If only he could enjoy the fruit of his labors.