Life in Voxels

In Episode 2 of Life in Voxels, we welcome Adam L. Kesner, PhD, Attending Physicist in the Department of Medical Physics at Memorial Sloan Kettering Cancer Center, for an extended conversation on the future of quantitative nuclear medicine and personalized radiopharmaceutical therapy.

If we can image and quantify where a therapeutic radiopharmaceutical goes inside an individual patient, why do we still often prescribe treatment primarily according to administered activity rather than patient-specific absorbed dose?

Together with hosts Nikolaos A. Karakatsanis, Georgios Soultanidis and Ezzat Elmoujarkach, Adam takes us from the fundamentals of quantitative PET and SPECT to the practical challenges of making clinical dosimetry accurate, reproducible and scalable.

We discuss the MIRD framework, patient-specific absorbed-dose estimation, quantitative imaging and partial-volume effects, pharmacokinetic modeling, uncertainty, clinical workflow barriers, and the MIRDsoft family of community dosimetry tools—including MIRDcalc, MIRDfit, MIRDy90 and MIRDpvc.

The conversation also explores the role of AI and automation, the need for standardized and interoperable dosimetry data, the design of radiopharmaceutical therapy clinical trials, and the provocative question of whether conservative treatment paradigms may leave some patients undertreated.

Adam also reflects on his earlier work in data-driven PET respiratory motion correction and what it taught him about an issue that runs throughout this episode: a technically elegant solution only changes medicine when it can be made practical enough to work in real clinical care.

This is a wide-ranging discussion about how nuclear medicine can move from measuring radiopharmaceutical biodistribution to using those measurements to guide therapy for the individual patient.

🎙️ Life in Voxels — conversations with the scientists, engineers, medical physicists and clinicians shaping the future of medical imaging.

How does a PET scanner progress from an initial scientific concept to a functioning imaging system?

In the inaugural episode of Life in Voxels, hosts Nicolas A. Karakatsanis, Georgios Soultanidis, and Ezzat Elmoujarkach are joined by instrumentation scientists Fiammetta Pagano of the Institute for Instrumentation in Molecular Imaging (i3M), Valencia, Spain, and David Sánchez of Oncovision SA, Valencia, Spain.

Their conversation follows the development of PET instrumentation from scientific need and detector design through system integration, calibration, simulation, and image reconstruction, using the Ultra-High-Performance Brain PET (UHB-PET) prototype as a real-world example.

The discussion explores why dedicated brain PET systems remain scientifically important; how semi-monolithic detector architectures work; why depth-of-interaction information matters in compact PET geometries; how machine learning can estimate gamma-ray interaction positions; how multiplexing helps translate laboratory detector concepts into scalable systems; and what changes when hundreds of detector components are integrated into a complete scanner.

The conversation also goes behind the scenes of UHB-PET calibration and commissioning at Weill Cornell Medicine, examining energy and timing calibration, detector positioning, normalization, system troubleshooting, experimentally informed Monte Carlo simulation, image reconstruction, and the recovery of complex detector events.

Finally, the guests discuss the increasingly close relationship between hardware and software innovation and what the next generation of PET instrumentation may look like.

Hosts

Nikolaos A. Karakatsanis — Department of Radiology, Weill Cornell Medicine
Georgios Soultanidis — Biomedical Engineering and Imaging Institute, Icahn School of Medicine at Mount Sinai
Ezzat Elmoujarkach — Department of Radiology, Weill Cornell Medicine

Guests

Fiammetta Pagano — Institute for Instrumentation in Molecular Imaging (i3M), Valencia, Spain
David Sánchez — Oncovision, Valencia, Spain

Topics Discussed

PET instrumentation • dedicated brain PET • UHB-PET • semi-monolithic detectors • LYSO scintillators • silicon photomultipliers (SiPMs) • depth of interaction (DOI) • time of flight (TOF) • machine learning • detector multiplexing • scanner calibration • system commissioning • GATE Monte Carlo simulation • image reconstruction • inter-crystal scattering • artificial intelligence in detector technology

Produced and published by POSITROPTICS LLC.

© 2026 POSITROPTICS LLC. All rights reserved.

The views expressed by the hosts and guests are their own and do not necessarily represent those of their respective institutions or organizations.