Before Zhengrong “Ron” Ying built CT scanners, he worked on a problem that sounds almost philosophical: how do you recognize an object when part of it is hidden? In 2001, as an electrical-engineering doctoral student at Boston University, Ying presented a poster called “A Bayesian Framework for Occluded Nonrigid Object Recognition.” The title is dry enough to preserve office plants. The question underneath it is not. The world rarely offers a perfect view. Something blocks the subject. The subject bends. The signal arrives incomplete. The useful machine must still decide what it is looking at.
That poster earned a Dean’s Award at the university’s Science Day. It also supplied an early thumbnail of the career that followed. Ying would spend the next two decades working on machines designed to see what ordinary sight cannot: first through research and patents, then through products, and eventually through a company of his own.
His route ran from Shanghai to Boston to Suzhou. Along it, the vocabulary expanded from Bayesian models and deformable contours to detectors, multi-energy X-rays, reconstruction, gantries and product lines. Yet the basic pursuit remained steady. Gather imperfect measurements. Turn them into a legible image. Make the image useful enough to guide a decision.
The question beneath the machinery
Ying studied electrical engineering at Shanghai Jiao Tong University from 1990 to 1997, completing bachelor’s and master’s degrees. He then went to Boston University for a doctorate in electrical engineering. His adviser was David Castañón, whose work sits at the intersection of signal processing, estimation and control. The fit makes sense in retrospect. Computed tomography is a cathedral built from those disciplines, only with more cables.
A CT image does not begin as a photograph. It is reconstructed from measurements gathered at different angles. Hardware produces the signal; mathematics interprets it; software turns it into a volume a human can inspect. Each layer can introduce error, and each correction can disturb another part of the chain. The resulting engineering problem is less like making a camera and more like conducting an orchestra whose musicians also designed their instruments.
“The work changes costume. The central curiosity remains.”The thread through Ying’s career
After Boston University, Ying worked in the United States on security-screening and medical CT systems. Bowing Medical’s biography describes him as a former chief scientist at Analogic, a Massachusetts company with a long history in imaging. His papers and patent record from those years show a researcher moving around the entire system rather than occupying one tidy corner.
There was work on detecting threat objects in CT scans, extracting three-dimensional baggage images from successive slices, correcting multi-energy measurements, changing scanning pitch and reconstructing the resulting data. A 2007 conference paper examined dual-energy volumetric X-ray tomography for luggage screening. A 2008 paper in IEEE Transactions on Medical Imaging tackled variable-pitch reconstruction using John’s equation. Different applications, same stubborn demand: derive a trustworthy view from complicated measurements.
From the suitcase to the scanner room
Airport baggage and medical imaging seem like unrelated neighborhoods. CT physics provides the bridge. In both, the system must distinguish materials and shapes hidden inside an object. In both, bad information has consequences. And in both, the scanner is only as useful as its weakest link, whether that link is the source, detector, calibration, reconstruction or interface.
Ying’s patent trail reads like a tour of those links. One invention concerns finding anomalies in projection images before they become streaks in the final scan. Another separates compound objects. Others address detector positioning, photon counting, multi-pixel X-ray sources, three-dimensional workstations and iterative reconstruction across raw, projection and image domains. The interesting number is not how many patents carry his name. It is how insistently they refuse to stay in one discipline.
That breadth later became a founder’s asset. Deep-technology companies are punished for departmental optimism. The algorithm cannot merely sparkle in a notebook. The detector has to collect the data, the gantry has to move, the reconstruction has to finish, the product has to be manufactured, and the operator has to understand the result. A founder who has worked across the seams carries a map of where apparently local decisions become system-wide problems.
One problem, four rooms
Suzhou as a design decision
In 2012, Ying and his partner Yu Li began a different kind of reconstruction. They co-founded Suzhou Bowing Medical Technology, also known as BOZWIN, in Suzhou Industrial Park. The founding team brought years of Boston-area imaging experience. The company established product development, manufacturing and sales in Suzhou while maintaining a research connection in the Boston area.
When Ying later explained the choice of Suzhou, he did not reach for the usual fog of ecosystem language. He pointed to two practical forces: local government services organized to help companies get things done, and a dense community of overseas returnees who created an entrepreneurial culture. Put enough capable people close together, he suggested, and “things can roll.” It is an engineer’s description of a city: inputs, proximity, momentum.
The company arrived in an industry with formidable barriers. A CT system combines thousands of physical components, extensive intellectual property and a regulatory path that does not reward theatrical improvisation. Bowing Medical began with core components and developed toward complete scanners. By 2019 it had introduced a high-end CT system. In 2020 it added vehicle-mounted and cabin-based systems while laying plans for intraoperative CT.
The product list matters because it reveals the company’s thesis. General-purpose imaging is one market; specialized settings create a set of narrower, harder questions. A mobile scanner has different spatial and operational constraints from a fixed installation. A system designed around a particular procedure must fit the procedure rather than demanding that the procedure fit it. Bowing Medical’s public vision phrases this progression as moving from precise diagnosis toward intelligent treatment.
The founder of the whole stack
There is a fashionable picture of technical founders as people who leap from insight to industry. Ying’s record looks more cumulative. The student poster led into papers. Papers lived beside patents. Patents accumulated across subsystems. Only then did the company become the container for the work. What appears, from a distance, to be one leap is better understood as stored context finding a larger job.
Recognition followed, though it arrives in the record with the bureaucratic neatness common to engineering careers. Public university biographies list Ying as a nationally specially appointed expert and a Gusu innovation and entrepreneurship leading talent. They also note an excellence award in a field round of China’s national disruptive-technology innovation competition and an intellectual-property financial innovation award. Bowing Medical identifies him as an IEEE Senior Member and a member of an international CT professional committee. Awards can flatten a career into nouns. Here, they are more useful as coordinates: research, invention, company building and the patient translation of one into the next.
The sequence offers a lesson for builders working on physical products. Expertise is not merely knowing more facts about one component. It is recognizing which interface will betray you next. Reconstruction depends on calibration. Calibration depends on geometry. Geometry depends on mechanics. Mechanics meet cost, manufacturing and service. In a system this connected, fluency at the boundaries is a competitive capability.
“Be honest; do practical work.”Bowing Medical’s stated values, translated from Chinese
That company maxim is unadorned to the point of charm. It also fits the tempo of Ying’s public career. He has maintained a relatively spare public profile while the work appears in patent documents, technical papers, university biographies and product announcements. The personality visible through those records is the personality of the systems engineer: patient with complexity, suspicious of isolated answers, interested in the full route from signal to decision.
In 2022, during a visit by radiology professor Xiao Xiangsheng, Ying spoke about continuing to invest in research while expanding beyond general CT into specialized and intelligent-treatment applications. In 2023, the University of Science and Technology of China listed him as an industry-education mentor. A 2025 patent filing named him on work concerning CT raw-data compression, recovery and storage. In 2026, he was listed to speak on the path from accurate diagnosis to intelligent treatment at a medical-electronics summit.
The details change: occluded shapes, baggage, detectors, raw data, operating rooms. The career keeps returning to the same elegant difficulty. Reality is hidden. Measurements are incomplete. The machine must assemble a view good enough for someone to act. Ying has spent decades improving that assembly, then building the organization around it. Seeing inside things, it turns out, is also a way of seeing a company.