Graduate Seminar: Shekhar Bhansali, Oct. 21, 2026

Summary

1. A deliberately interdisciplinary career

The seminar speaker describes a career spanning metallurgy, aerospace, applied physics, electrical engineering, MEMS, microfluidics, biosensing, and additive manufacturing. His research direction changed as new questions and practical constraints emerged. For example, when a clean room at one institution was largely unusable, he shifted toward applied and agricultural sensing. His central lesson is that a research career may seem like a series of pivots at the time, but its themes often connect in hindsight.

He encourages students to pursue interdisciplinary, open-ended problems, even when they cannot predict where the work will lead. He also contrasts this work with repetitive production-line engineering, which he found less satisfying.

2. Sensors for continuous measurement

A recurring goal in his work has been to create sensors that collect data continuously, including wearable devices. He describes work on microfluidic devices, MEMS, magnetic-bead sensing, nanomaterials, and sensors for physical, chemical, and biological measurements.

For health sensing, he emphasizes that different modalities and body fluids offer different possibilities and challenges:

• Modalities: Biochemical, electrical, optical, and acoustic measurements.
• Potential fluids: Sweat, saliva, tears, blood, and interstitial fluid.
• Practical constraints: Fluid accessibility, analyte concentration, sensor specificity, stability, cost, and ease of deployment.
• Translation: A device that works in a laboratory may still be difficult to commercialize or use reliably outside it.

He notes that choosing a measurable target requires understanding what is already known about its concentration and behavior in different fluids.

3. Cortisol sensing and wearable health monitoring

The speaker discusses cortisol as a target because it can be measured in body fluids and is relevant to stress, PTSD, and depression research. His group explored ways to make sensors more stable and potentially reusable:

• They used single-domain antibodies derived from camelids, which he described as more heat-stable than conventional antibodies.
• They also investigated molecularly imprinted polymers: polymer layers formed around a target molecule so that, after the target is removed, the layer retains a matching impression.
• Applying voltage could help release captured molecules, but the polymer layer degraded with repeated use. It worked for a small number of cycles, then began peeling off with more extensive cycling.
• His group compared sensor performance across body fluids and against commercial ELISA tests. They also used cortisol sensors in studies involving pregnancy and depression.

During the talk, he uses a brief mental-arithmetic exercise to illustrate how quickly stress can affect cortisol. He adds that, in depression research, a collaborator was interested not only in how high cortisol rises, but also in how it falls.

4. Wound sensing: choosing markers and models carefully

The group later pursued wearable sensing for wound care. The speaker describes focusing on lactate and uric acid, using enzymatic sensing approaches. Uric acid can be associated with cell death, but it is not specific to wounds and can vary for other reasons.

To explore whether the measurements could distinguish wound conditions, the team compared material collected from wounds, nearby skin, and healthy skin. The speaker stresses an important research-design lesson: a chronic wound may be a poor model for validating a sensor intended to track healing, because a chronic wound is defined by its failure to heal. A healing surgical wound may provide a clearer changing signal.

Early wearable versions did not produce useful continuous wireless data. The team redesigned the electronics and tried an SD-card version, but the results remained inconclusive. COVID-era constraints also limited further testing.

5. Commercialization and patents

A separate example involved a fuel-cell-based wearable alcohol sensor. Although the device generated interest, the team encountered the difficulty of getting such a product approved across different jurisdictions. A commercial competitor also introduced a design that, in the speaker’s account, avoided the team’s patent.

His takeaway is that commercialization is a distinct challenge from demonstrating a working prototype. He advises researchers to consider broad patent claims rather than writing a patent that merely mirrors the details of a research paper.

6. Additive manufacturing, sensing, and AI

The speaker connects earlier work on cell cultures and tumor spheroids to his current interest in additive manufacturing. He had wanted to measure tumor spheroids from multiple directions but lacked a practical way to make suitable three-dimensional electrode systems. Additive manufacturing offered a way to create more complex structures, including porous scaffolds intended for bone regeneration.

His current research direction includes monitoring the manufacturing process itself:

• Integrate sensors into a 3D printer to measure conditions such as temperature and humidity.
• Combine process data with images and micro-CT measurements of printed parts.
• Use the combined data to identify defects and predict what may happen during printing.
• Address limited AI training data by using techniques such as transfer learning and generating additional image examples through small rotations or other transformations.

He also describes exploring collaborations between electrochemical sensing and photonics, combining measurement approaches that had previously been developed in separate research areas.

7. Advice for students and researchers

The speaker’s practical lessons include:

• Expect research directions to change. Constraints and unexpected results can lead to productive pivots.
• Choose a model that can test the question. A model that does not change in the relevant way may not validate a sensor effectively.
• Plan for translation early. Cost, stability, usability, and deployment matter alongside laboratory performance.
• Treat collaboration as essential. Many of the projects described depended on expertise from outside his own field.
• Apply for the NSF Graduate Research Fellowship early. He advises undergraduates to apply while still eligible as undergraduates, since they may have another opportunity as graduate students.
• Consider whether the work fits your interests. He found repetitive production work less fulfilling than research involving new, unresolved problems.

Speakers and sources mentioned

Speakers

• Dr. Kim — Introduced the seminar speaker.
• Professor Shekhar Bhansali — Seminar speaker; his surname is rendered as “Bansal” in parts of the auto-generated subtitles.
• Unidentified audience members — Responded during the cortisol demonstration and asked a closing question.

People, organizations, and research examples referenced

• Erwin Schrödinger and his book What Is Life? (1944), cited in connection with interdisciplinary research.
• NSF, including the NSF Graduate Research Fellowship Program and research-funding programs.
• NIH, mentioned in relation to an earlier grant proposal.
• DOE and DOE-affiliated facilities or contractors, including Y-12, Honeywell, and KCNSC.
• Stanford’s Snyder lab, mentioned in connection with personalized health data and startups.
• Dexcom, FreeStyle Libre, and ELISA tests, discussed as examples of existing health-monitoring or measurement technologies.
• Collaborators and colleagues named in passing include Mark, Cynthia, Vivek, Karthik, Kshitij, and Veronica. The subtitles do not provide full identifying details for them.