During the COVID pandemic, when Brazil’s Ministry of Health called on researchers to validate and calibrate ventilators intended for hospitals in the Unified Health System (SUS), one group encountered a gap in the global market for appropriate testing equipment. To determine whether the ventilators were safe, the researchers had to test them in simulators that replicated human respiratory mechanics. Most commercially available simulators, however, are sophisticated and expensive devices designed for adults and older children. None fully met the criteria required by the international standards for critically ill premature infants.

Neonatal ventilation presents a physical challenge. A very low-birth weight premature infant requires tidal volumes measured in mL, delivered against a stiff and immature lung parenchyma. The regulatory standard ABNT NBR IEC 80601-2-12 requires neonatal simulators to have a compliance of 0.5 mL/hPa, whereas commercially available devices reach a maximum of 1.0 mL/hPa.

To address this gap, biomedical engineer Sara Gomes Mello Santiago and her adviser, Henrique Takachi Moriya, PhD, associate professor in the Department of Telecommunications and Control Engineering at the Polytechnic School, University of São Paulo (USP), São Paulo, Brazil, partnered with the Institute for Technological Research to develop six low-cost neonatal simulator designs. The solution used commercially available 50-mL and 20-mL silicone bellows as raw materials, together with an adjustable containment system made from acrylic plates and steel coil springs. One configuration achieved a static compliance of 0.52 mL/hPa and was approved in Inmetro audits for the accreditation of testing laboratories in Brazil.

Speaking with Medscape’s Portuguese edition, Santiago said, “The key aspect of Biodesign is language. When we talk to doctors and physical therapists at the hospital, they use everyday clinical terms and experience firsthand the lack of a tool. However, they do not know how to translate that need into engineering development. Biomedical engineering acts as a transducer: It takes the problem presented by the healthcare team and translates it into technological parameters, while also interpreting regulatory standards and creating a practical, reproducible solution for the day-to-day work of clinical engineering in hospitals.”

The transducer converts energy without losing information. This principle underlies biodesign, a health innovation methodology that trains doctors, engineers, and managers to work across disciplinary boundaries. In Brazilian laboratories, clinics, and hospitals, researchers have learned to communicate across disciplines that have previously struggled to understand one another.

Translation Skills

For many years, medical innovation has been associated with innate talent. Either someone had an eye to innovation or did not. The Stanford Mussallem Center for Biodesign in Stanford, California, challenged this idea by turning innovation into a process based on the Identify, Invent, and Implement cycle that can be taught and practiced.

“The big change is training doctors and engineers to spend much more time understanding the real problem than trying to invent a product too early,” said Robson Capasso, MD, a professor of Otolaryngology - Head & Neck Surgery and Program Development Director, Stanford Mussallem Center for Biodesign.

This shift can be substantial for doctors trained to make accurate diagnoses. “Doctors are trained to be certain and not to make mistakes,” said Rodrigo Bornhausen Demarch, MD, Chief Innovation Officer at Hospital Israelita Albert Einstein, São Paulo.

Demarch was the first Brazilian accepted into the Stanford Medicine Global Faculty Training Program (formerly Global Faculty in Training) in 2017 and led the introduction and scaling of Biodesign-inspired programmes in Brazil.

“In innovation, this mindset needs to change. Professionals need to adopt the mindset of a learner, quickly test hypotheses, discard what doesn’t work, and rebuild on the basis of what they’ve learned,” he said.

Einstein’s 8-month fellowship requires 20 hours per week. Participants completed internships at private and public hospitals and worked in teams comprising a physician, an engineer, and management or business professionals. “The doctor specializes in clinical pain, the engineer knows how to build and prototype, the business professional understands the business model and how to bring the technology to market,” Demarch said.

Biodesign is not limited to clinicians who wish to establish companies. The program also enables clinicians to approach patient care through continuous optimization, improving day-to-day care, whether or not they go on to create a startup or file a patent.

Three quarters of medical technology startups do not provide a return on investment. Most do not fail because of engineering problems. Instead, they failed because the key questions were not addressed: Will the payer reimburse the cost? Will the regulator approve the product? Will nursing staff use it? Can the emergency department workflow accommodate this?

“If you haven’t understood the problem in depth, you don’t know whether the solution makes sense for the hospital’s workflow or for the payer’s bottom line,” Demarch said. The local context adds another layer of complexity. “The way reimbursement works in the SUS and in private health insurance makes all the difference when it comes to determining whether a problem is worth solving,” he added.

Conrado Tramontini, manager of the Innovation Garage at Hospital Sírio-Libanês in São Paulo, put it this way: Innovation works “when we have a real problem seeking a solution, and never a technology seeking a problem.”

Biological Barriers

There are cases in which the obstacle is neither regulatory nor financial. The drug exists, the patient exists, and a diagnosis exists. What is missing is the path forward.

Glioblastoma multiforme has a median survival of less than 2 years. Numerous drugs can kill tumor cells in laboratory studies, but delivering these drugs to the tumor remains a challenge. The blood-brain barrier prevents more than 98% of the therapeutic agents administered orally or intravenously from reaching the tumor. Temozolomide, the standard chemotherapy for this disease, has a half-life of only 2 hours in the blood before it is inactivated. This necessitates high doses with severe adverse effects so that even a small fraction of the drug can cross the barrier.

Researcher Natália Noronha Ferreira Naddeo, PharmD, PhD, a postdoctoral researcher in the Nanomedicine and Nanotoxicology Group (GNano) at the São Carlos Institute of Physics (IFSC), USP, was studying nasal drug delivery for another indication when a question during her doctoral defense redirected her research. A glioblastoma specialist asked why the nasal route, if promising, could not be used to deliver temozolomide, a drug that is difficult to deliver effectively in clinical practice.

Working with Valtencir Zucolotto, PhD, professor and coordinator, GNano, IFSC USP, and researchers from Hospital de Amor, formerly known as the Barretos Cancer Hospital in Barretos, and the Federal University of Goiás, Goiânia, both in Brazil, Naddeo developed polymeric nanoparticles loaded with temozolomide and coated with fragments of the tumor’s own cell membrane.

“By coating the nanoparticle with the tumor’s own membrane, we give it homologous adhesion,” she said. “Surface proteins enable the nanoparticle to recognize and preferentially adhere to tumor cells, while ignoring healthy brain tissue.” Administered nasally, nanoparticles reach the brain through the olfactory and trigeminal nerves, bypassing the blood-brain barrier.

The study, published in ACS Applied Materials & Interfaces in December 2024, showed promising results in tests involving tumor cells and animal models. The nanoparticles reached the brain within 30 minutes after intranasal administration, reduced cancer cell viability, and inhibited the blood vessels supplying the tumor.

This research won first place in the Technological Innovation in Oncology category at the 17th Octavio Frias de Oliveira Award 2026, awarded by the São Paulo State Cancer Institute, São Paulo. The next step is to consolidate in vivo preclinical studies and address the challenges of large-scale production.

This platform may have applications beyond glioblastomas. “We are already adapting this same strategy for breast and pancreatic tumors and using the nasal route to treat spinal muscular atrophy,” Zucolotto said.

High-risk non-muscle-invasive bladder cancer affects approximately 13,000 individuals annually in Brazil. The standard treatment after surgical resection is intravesical Bacillus Calmette-Guérin (BCG) therapy. BCG fails in approximately half of the patients. When this occurs after two treatment cycles, radical cystectomy is indicated, with complication rates from 60% to 90% and permanent use of a collection pouch.

Many patients are older adults and are not medically fit for surgery. Others simply refused the procedure. A global shortage of oncologic BCG, which prompted the American Urological Association to issue formal warnings and the US Food and Drug Administration to authorize alternative formulations on an emergency basis, has made the search for alternatives a priority beyond Brazil.

Wagner José Fávaro, PhD, associate professor of Human Anatomy in the Department of Structural and Functional Biology from the Institute of Biology, University of Campinas (Unicamp), Campinas, Brazil, developed OncoTherad, a synthetic nanocomplex designed to stimulate the immune system to attack tumors.

In a phase 1/2, open-label, single-arm clinical trial at Paulínia Municipal Hospital, Paulínia, Brazil, involving 44 patients with an indication for radical cystectomy, OncoTherad achieved complete tumor elimination in 72.7% of patients. No patient required bladder removal, and no deaths occurred during the 2-year follow-up period.

The adverse effects were mild and included dysuria, cystitis, joint pain, and low-grade fever. To understand why a smaller group of patients did not respond, researchers sequenced tumor ribonucleic acid and identified a resistance mechanism. An enzyme called monoamine oxidase A (MAO-A) was overexpressed in the tumor microenvironment and suppressed the lymphocytes that should attack the cancer.

The finding is consistent with independent research from the University of California in the US identifying MAO-A as a critical regulator of tumor immunosuppression across multiple cancer types.

This discovery led to the development of ImmunoClor, a second-generation platform that combines nanocomplexes with MAO-A inhibitors. MAO inhibitors have been approved for decades for use in neurology and psychiatry. ImmunoClor repurposes an existing class of drugs. “Imagine a bowling ball,” Fávaro said.

“It has a certain weight and occupies a concentrated space. Now, imagine eight billiard balls with the same total weight. The area they cover is infinitely large. The nanoparticle does this: We use a very small amount of the drug, reducing adverse effects, but cover a much larger area of contact with the tumor,” he explained.

Multicenter clinical trials of ImmunoClor are underway at hospitals in São Paulo. To bring technology to the market, the researchers established Aeter Biotech, an academic spin-off and university subsidiary of Unicamp, one of Brazil’s leading public research universities, and licensed the technology through Inova Unicamp, the university’s innovation and technology transfer agency.

Light-Based Therapy

Photodynamic and photobiomodulation therapies use light to inactivate bacteria and modulate inflammatory processes in the lungs. These noninvasive approaches apply light externally to the chest. The challenge is that light must pass through the skin, fat, muscle, and bone before reaching the lungs. Because anatomy varies among individuals and pneumonia can alter tissue density, applying the same light dose to every patient is imprecise.

Researchers at the IFSC USP working with researchers from Texas A&M University in College Station, Texas, and the Victoria University of Wellington in Wellington, New Zealand, used CT scans of patients to create detailed 3D models of the chest. The simulation software calculated the path of millions of photons through the rib cage.

The study, led by Cristina Kurachi, PhD, from the Physics and Materials Science Department, and Vanderlei Salvador Bagnato, PhD, professor of physics, both from IFSC USP, published in Proceedings of the National Academy of Sciences in August 2026, showed that areas of the lung affected by pneumonia absorb between 3.0% and 4.5% of the light energy, proportionally more than healthy parenchyma because they are denser. Therefore, light becomes concentrated where the disease is present.

The model allows clinicians to plan the optimal light configuration virtually before activating the device without an invasive procedure. “We’ve moved beyond simply illuminating without knowing the exact outcome to a personalized planning approach,” Kurachi said.

According to Bagnato, the next step is an in vivo study in sheep in the US in preparation for human trials.

At the Innovation Garage at Hospital Sírio-Libanês, the same principle is being applied to clinical data. BRONCOAlert is an automated clinical surveillance and early warning system designed to predict and prevent bronchoaspiration. By continuously analyzing data directly from electronic health records, the system works dynamically to identify at-risk hospitalized patients before an adverse event occurs. On another front, thoracic surgeons at the hospital have begun using mixed-reality glasses to view tumors in 3D during surgery.

In both cases, the researchers did not create new information. Instead, they asked questions that had not been previously asked.

Translation Barriers

The biological barriers encountered in earlier studies had precise locations in the human body. Barriers within the healthcare system are less defined but equally real.

“Ideas don’t fail because the engineering was wrong, but because the need was never validated within the ecosystem,” Capasso said. Brazil’s healthcare ecosystem has its own challenges, including regulations that may not yet address new technologies, a reimbursement system divided between the SUS and private healthcare, and longstanding dependence on imported pharmaceutical supplies.

OncoTherad illustrated this second barrier to implementation. Developing technology from the laboratory to phase 1 costs $7.9 million, almost 10 times less than the US benchmark of $74.6 million. Unicamp holds 100% of the patents in Brazil, the US, and Europe. However, bringing technology to the market remains difficult.

Brazil’s National Health Surveillance Agency (Anvisa) does not have good manufacturing practice regulations for nanomedicines, preventing the certification of a production facility at scale.

“To keep the science moving, we created Regional Innovation Hubs: The University of Campinas provides the technology, while municipalities provide hospitals to treat patients in the SUS without requiring direct financial transfers,” Fávaro said. The approach addresses an immediate need but also illustrates the challenges researchers face when the infrastructure needed to move innovation forward is not yet established.

In 2026, Anvisa launched a call for accelerated regulatory assessment of innovative medical devices and identified regulatory efficiency as an institutional priority. Therefore, the system barrier was changing, although the source noted that regulatory development had not kept pace with emerging technologies.

Roger Chammas, MD, PhD, professor of oncology at the USP School of Medicine and coordinator of the MBA in Health Innovation Management at Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, described the effect of bringing professionals from different disciplines into healthcare.

“When a physicist enters the hospital with an analytical perspective or a communications professional asks a question that a specialist had stopped asking, they challenge us to be not only better professionals but better human beings,” he said.

The MBA brings together Hospital das Clínicas, InovaHC, its technology and innovation center, and the Butantan Institute, a premier biomedical research and vaccine manufacturing center, all based in São Paulo, with the aim of changing how professionals view the relationship between scientific advances and patient needs.

Technology is not meant to replace physicians. Instead, it can help restore something that medicine has lost amid the system’s bureaucracy: time for patient care.

“If I don’t have a firm grasp of AI, I need to have the humility to learn,” Chammas said. “The true beauty will lie in reclaiming our essence — making time to listen to and welcome patients.”

What Santiago described as a “transducer” is ultimately a way of translating a clinical problem into a technical parameter and then into a solution for the patient. Biological, physical, and systemic barriers can become more manageable when professionals learn to communicate across disciplines.

Daniela Barros is a journalist from Brazil, specializing in social journalism at the Pontifical Catholic University of São Paulo, and a master’s student in the Department of Social Medicine at the Ribeirão Preto Medical School in Ribeirão Preto, Brazil. She has been involved in medicine for 23 years and has contributed to several specialized publications. 

This story was translated from Medscape’s Portuguese edition.