Beyond Anatomical Imaging: The Rise of Quantitative Tissue Characterization with Ultrasound
By Md Ashikuzzaman, Ph.D., Assistant Professor, Director, Medical Imaging for Global Health Technology (MIGHTY) Lab, Electrical and Computer Engineering, School of Science and Engineering, University of Missouri-Kansas City (UMKC)
Ultrasound has earned its place as one of the most widely used imaging technologies in medicine. It is safe, portable, relatively inexpensive, and capable of producing real-time images at the patient’s bedside. From obstetrics and cardiology to emergency medicine and cancer diagnosis, it has become an indispensable part of modern healthcare. Yet despite decades of remarkable progress, conventional ultrasound still answers only one part of the clinical question: What does the tissue look like?
Increasingly, clinicians want to know something more. How does the tissue behave? Is it becoming stiffer because of fibrosis? Is a suspicious lesion mechanically different from the surrounding tissue? Is a treatment changing tissue properties before visible anatomical changes appear? These questions cannot always be answered by appearance alone.
The next chapter of ultrasound is being written around this very idea. Ultrasound is becoming a tool that quantitatively assesses tissue qualities rather than just producing visuals. This transition has the potential to make diagnoses more objective, improve disease monitoring, and provide clinicians with information that was previously beyond the reach of conventional imaging.
One of the clearest examples is ultrasound elastography, which estimates the mechanical properties of tissue by measuring its response to an applied force or naturally generated shear waves. Tissue stiffness has become a valuable biomarker in many clinical applications. Liver elastography is now routinely used to assess fibrosis, substantially reducing the need for biopsy in many patients. Breast, thyroid, and prostate imaging have also benefited from stiffness measurements that complement conventional B-mode ultrasound. Instead of relying solely on visual interpretation, clinicians gain access to numerical information that reflects how tissue behaves under mechanical loading.
The future of Ultrasound is therefore not simply about producing sharper images. It is about extracting meaningful measurements that reveal how tissue functions, changes, and responds to disease or therapy.
Mechanical properties, however, are only one piece of the puzzle. Ultrasound can also quantify characteristics such as blood flow, tissue perfusion, attenuation, and acoustic scattering. These measurements provide insights into tissue composition and function that extend beyond anatomy. As a whole, they move ultrasound closer to what clinicians have long wanted: objective biomarkers that support diagnosis, treatment planning, and longitudinal monitoring.
This shift is important because many diseases begin with changes that are difficult to see but easier to measure. A liver affected by early fibrosis may still appear relatively normal on conventional imaging while its mechanical properties are already changing. Likewise, two lesions may look remarkably similar on a grayscale image yet exhibit very different stiffness or vascular characteristics. In such cases, quantitative measurements add another layer of clinical information that complements visual assessment rather than replacing it.
Of course, measuring tissue is considerably more difficult than taking a picture. Biological tissue is constantly moving. Patients breathe. Blood vessels pulsate. Operators apply different amounts of pressure during handheld scanning. Even small changes in probe position can influence measurements. Quantitative ultrasound is more analogous to attempting to gauge an object’s firmness (or other fundamental properties) while both your hand and the object are moving, whereas conventional ultrasound is more akin to capturing a picture. The challenge is substantial, which explains why reliable quantitative imaging has remained an active area of research for many years.
Recent advances in artificial intelligence are helping address some of these challenges. Much of the public discussion around AI in medical imaging focuses on disease classification, but its impact extends much further. AI is improving image reconstruction, automating anatomical measurements, enhancing image quality, and enabling more robust estimation of quantitative imaging biomarkers. These advances have the potential to make quantitative ultrasound faster, more consistent, and less dependent on operator experience.
At the same time, enthusiasm for AI should be matched with scientific rigor. Medical decisions require technologies that are reliable, explainable, and thoroughly validated across diverse patient populations. Models trained under carefully controlled conditions must also perform consistently in busy clinical environments where image quality, patient anatomy, and scanning protocols vary considerably. The goal should not be to replace clinical expertise with algorithms. Instead, AI should work alongside clinicians, helping extract meaningful information while preserving transparency and confidence in the measurements.
Another promising direction is the growing integration of physics with machine learning. Unlike conventional computer vision tasks, ultrasound images are governed by well-understood physical principles, including wave propagation, tissue mechanics, and acoustic interactions. Incorporating this knowledge into AI models offers an opportunity to improve robustness, reduce dependence on large, annotated datasets, and produce measurements that remain consistent with the underlying physics. Rather than viewing physics and AI as competing approaches, the future likely belongs to methods that combine the strengths of both.
Looking ahead, quantitative tissue characterization is likely to become a routine component of ultrasound examinations rather than a specialized add-on. Portable ultrasound systems are becoming increasingly powerful, computational hardware continues to improve, and advanced algorithms are moving closer to real-time performance. These developments could make quantitative imaging available not only in major medical centers but also in community hospitals, outpatient clinics, and resource-limited settings where ultrasound already plays a critical role.
The future of ultrasound is therefore not simply about producing sharper images. It is about extracting meaningful measurements that reveal how tissue functions, changes, and responds to disease or therapy. Images will always remain central to clinical practice, but numbers that describe tissue mechanics, composition, and physiology will increasingly accompany them. Together, they promise a more complete picture of human health.
For decades, ultrasound has allowed clinicians to see inside the body. That achievement transformed medicine. Teaching ultrasound to measure tissue with the same confidence may prove to be its next defining milestone.
References
1. Tsuyoshi Shiina et al., WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 1: basic principles and terminology, Ultrasound in medicine & biology, 2015.
2. Christoph F Dietrich et al., EFSUMB guidelines and recommendations on the clinical use of liver ultrasound elastography, update 2017 (long version), Ultraschall in der Medizin-European Journal of Ultrasound, 2017.
3. Thomas L Szabo, Diagnostic ultrasound imaging: inside out, Academic press, 2013.
4. Eric Topol, Deep medicine: how artificial intelligence can make healthcare human again, Hachette UK, 2019.
5. Guy Cloutier et al., Quantitative ultrasound imaging of soft biological tissues: a primer for radiologists and medical physicists, Insights into Imaging, 2021.
6. Zhuhuang Zhou et al., Recent advances in artificial intelligence-empowered ultrasound tissue characterization for disease diagnosis, intervention guidance, and therapy monitoring, Frontiers in Physiology, 2023.
7. Md Ashikuzzaman et al., Displacement tracking techniques in ultrasound elastography: From cross correlation to deep learning, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2024.
8. Christoph F Dietrich et al., History of Ultrasound in Medicine from its birth to date (2022), on occasion of the 50 Years Anniversary of EFSUMB. A publication of the European Federation of Societies for Ultrasound In Medicine and Biology (EFSUMB), designed to record the historical development of medical ultrasound, Medical Ultrasonography, 2022.
9. Hongliang Li et al., Deep learning in ultrasound elastography imaging: A review, Medical Physics, 2022.
10. Hendra Lo et al., Handheld ultrasound (HHUS): potential for home palliative care, Ultrasound International Open, 2022.