TY - JOUR
AU - Grodzinsky, Alan J.
AB - Abstract The most visible contributions of biomedical engineering to clinical practice involve instrumentation for diagnosis, therapy, and rehabilitation. Cell and tissue engineering also have emerged as clinical realities. In the next 25 years, advances in electronics, optics, materials, and miniaturization will accelerate development of more sophisticated devices for diagnosis and therapy, such as imaging and virtual surgery. The emerging new field of bioengineering—engineering based in the science of molecular cell biology—will greatly expand the scope of biomedical engineering to tackle challenges in molecular and genomic medicine. Biomedical engineering is broadly defined as the application of engineering principles to problems in clinical medicine and surgery. The origins of biomedical engineering are often linked to the pioneering electrophysiology studies of Galvani and Volta more than 200 years ago.1 During the first half of the 20th century, the electrical properties of tissues and cells continued to be a primary focus of biomedical engineering. Emerging interests in the health effects of ionizing radiation between World Wars I and II laid the groundwork for current radiation therapy. It was not until the 1960s and 1970s, however, that an explosion in multidisciplinary research, combining mechanical, chemical, and electrical engineering with physiology and medicine paved the way for dramatic advances in modern health care based on breakthrough discoveries in biomedical engineering.2 Clinical and Research Advances The most visible contributions of biomedical engineering to clinical practice involve instrumentation for diagnosis, therapy, and rehabilitation.3 A revolution in disease diagnosis began in the 1970s with the introduction of computed tomography, magnetic resonance imaging, and ultrasonic imaging.4-8 Computed tomographic scanning, developed at EMI Research Labs (Hayes, Middlesex, England) (funded in part by the success of EMI's Beatles records) provided the first high-quality, cross-sectional images of the body.9 Magnetic resonance imaging further revolutionized imaging of soft tissues, enabling both the imaging of static structures and dynamic function, such as blood flow and metabolism. Open magnetic resonance imaging scanners are used in operating rooms to guide biopsies and stereotactic surgery.7 Ultrasound also affords soft-tissue imaging, with lower resolution but at reduced cost and thus increased accessibility compared with magnetic resonance imaging.8 The development of these imaging modalities has been accompanied by exciting advances in 3-dimensional image reconstruction, quantitative image analysis, and image enhancement—advances that were made possible by improved computational power and algorithms. The development in the 1980s of the noninvasive pulse oximeter was a significant advance in intraoperative monitoring and postsurgical care.10 Biomedical engineering has also been responsible for the development of new therapeutic devices. The cochlear implant, for example, has now been used in 20 000 of the 1 to 2 million deaf individuals in the United States, many showing dramatic improvement.11 Sound is decomposed into critical frequency bands and signals are delivered electronically to auditory neurons via an array of electrodes. Cardiovascular therapy has been similarly changed by the introduction of lifesaving implantable defibrillators (1980s),12 and ventricular-assist and catheter-based ablation devices. In addition, vascular stent technology for the treatment of aneurysms, peripheral vascular disease, and coronary artery disease has made it possible for a minimally invasive procedure to replace major surgery.13 The shift toward minimally invasive surgery, driven by development of miniaturized cameras, laser guides, and surgical tools, has had a profound effect in reducing morbidity and mortality, and enabling faster rehabilitation. Cell and tissue engineering also have emerged as clinical realities. Products for skin replacement are in clinical use14,15 and progress has been made in developing technologies for repair of cartilage (Figure 1),16,17 bone, liver, kidney, skeletal muscle, blood vessels, the nervous system, and urological disorders.15 The discovery that mechanical forces are potent regulators of cell-mediated growth, degradation, and repair of musculoskeletal17,18 and cardiovascular19 tissues presents a major challenge and elucidation of the mechanisms involved in mechanical signaling and cellular mechanotransduction will be needed to develop new therapies for a true understanding of osteoarthritis, osteoporosis, and atherosclerosis, and for the development of tissue engineering therapies. At the same time, the field of biomedical engineering is undergoing a major ideological change. The fusion of engineering with molecular cell biology is pushing the evolution of a new engineering discipline termed bioengineering to tackle the challenges of molecular and genomic medicine. In much the same way that the iron lung (an engineered device) was rendered obsolete by the polio vaccine (molecular medicine), many of the device-based and instrumentation-based therapies in clinical use today will likely be replaced by molecular- and cellular-based therapies during the next 25 years. Realization of these therapies will require major contributions from bioengineering. Arguably the most well-recognized contributions of bioengineering have been at the level of obtaining molecular information—techniques that make it possible at the research, development, and even clinical levels to manipulate, sequence, reconstruct, and model proteins and nucleic acids, such as polymerase chain reaction, DNA sequencing, molecular modeling, and bioinformatics. On this molecular level, the most stunning successes in terms of data acquisition are clearly in genomics. For example, DNA microarrays, involving analysis of expression profiles of thousands of genes simultaneously, are being applied for biomedical research and pharmaceutical development.20 Yet many engineering challenges remain to be solved before the power of DNA microarray analysis is optimal.21-23 Many drugs developed through molecular-level assays prove to be ineffective, and bioengineering analysis is beginning to point the way to better drug design and better drug testing. Most pressing for genomics is the lack of appropriate computational tools for the analysis and interpretation of expression profile data (ie, bioinformatics). Current tools consist largely of algorithms that group genes according to shared expression patterns.21 In addition, hybridization is not always sufficiently reliable or sensitive; there is a great need for better design of hybridization chips as well as algorithms to account for data imperfections.22 The cellular messenger RNA levels measured by microarray analysis do not strongly correlate with corresponding protein levels, further clouding the interpretation of microarray data.23 Proteins are the molecules that regulate metabolic processes and signaling pathways and most drugs target proteins and protein-protein interactions.24 Thus, analysis of cellular protein profiles (proteomics) may be particularly valuable in drug discovery and analysis of disease states. Proteomics is less advanced than genomics because of the greater complexity of protein, but advances in mass spectrometry analysis of proteins are causing a revolution in proteomics.24 Proteomic analysis shares some of the same challenges as genomic analysis in terms of improving sample preparation, throughput, reliability, and means of analyzing massive data sets. Engineering analysis often entails building complex systems models by integrating detailed mathematical models of the physical and kinetic properties of the component parts. Such analysis entails making assumptions about the nature of missing information to build predictive models of the system as a whole when the properties of 1 component are altered. Even though the individual processes contributing to performance may be poorly understood, models of biological systems contribute to application of the biological information, such as improved, targeted drug therapies, in the absence of complete data on cellular genes and proteins. Insights gained from such complex models are often both nonintuitive and valuable. For example, high-throughput screening assays to identify compounds for potential drug development typically identify compounds that bind with high affinity to a receptor. Models of the cellular pharmacology of receptor-ligand interactions, which integrate separate mathematical descriptions of ligand binding and release, receptor activation and desensitization, and receptor internalization, recycling, and degradation indicate that classic strategies based solely on high-affinity equilibrium binding can fail to predict drug candidates—for some receptors, lower affinity compounds may be more effective. (Figure A) For example, transforming growth factor α is more potent than epidermal growth factor in stimulating cell growth, even though transforming growth factor α has a lower binding affinity than epidermal growth factor to their common cell surface receptor. Surprisingly, a detailed cellular bioengineering model shows that lower affinity binding, particularly at the pH of the endosome, leads to less degradation both of receptor and ligand and increases receptor recycling. More important, a ligand designed via molecular engineering (ie, rational design of protein-protein interactions), which incorporates a single amino acid change resulting in a binding affinity below that of epidermal growth factor or transforming growth factor α, is a more potent ligand.25 This type of cell-level analysis, based on defining interlinked dynamic processes, is now being applied to engineer drugs that affect other receptor pathways, such as interleukin 225 and G-protein-coupled receptors.26 The potential implications for the drug discovery process are profound. Such analyses and design principles are also being applied to tailor vectors for gene therapy to increase specificity and efficiency of transfer and to design synthetic molecules to control cell behaviors in tissue engineering.26 Cell- and molecular-level models are the building blocks for development of mathematical models of entire systems. The advent of powerful computers and graphical interfaces has made possible the development of complex models that span levels from genes to organ systems, including feedback responses across all these levels.28 Such models make it possible to predict how a complex, nonlinear, and expensive experimental system (eg, a nonhuman primate or a human) will respond based on input from cell culture or biochemical assays. In the next 25 years, several large-scale systems modeling efforts at the academic level (eg, the Physiome Project, http://www.physiome.org), and at the commercial level (eg, Physiome Sciences [Princeton, NJ] and Entelos Inc [Menlo Park, Calif]) will become available. For example, a simulation of asthma (Figure 2) that relates dynamic changes at the biochemical level (eg, leukotriene production) to functional outcomes (eg, the degree of compromised breathing in a patient with asthma)28 illustrates the utility of this approach. Forecast for Research Advances In the next 25 years, advances in electronics, optics, materials, and miniaturization will push development of more sophisticated devices for diagnosis and therapy, such as imaging and virtual surgery. The accelerating pace of development of bioMEMS (biomicroelectromechanical systems, integrating electrical, mechanical, and optical systems on a micro scale) and microfluidic (incorporating microlevel fluid pumping, mixing, and reaction circuits) systems, combined with bioinformatics, will likely give rise to a new era of "lab on a chip" diagnostics, enabling routine and sensitive analysis of thousands of molecules simultaneously from a single sample. Such analysis might be done on a yearly basis in the way cholesterol screening is now done. (Figure B) A potentially even greater impact of bioengineering will result from the increased ability to incorporate molecular-level information into complex models. The result will be a revolution in diagnosis and treatment of diseases ranging from osteoarthritis to Alzheimer disease. Either by looking for single-signature molecules (eg, cancer antigens) or by using appropriate algorithms to derive relationships between many interacting molecules, early prediction of onset of disease may be possible. For example, osteoarthritis might be detected just when cartilage degradation begins and before damage is irreversible; Alzheimer disease might be detected in early adulthood when it is believed lesions might first form and before cognitive decline. In each case, new drugs developed with the aid of molecular and cellular engineering will likely be available to combat disease progression. For osteoarthritis, these advances would obviate the need for joint replacement surgery or even for cell transplantation. For Alzheimer disease, which lacks current therapeutic options, the impact of bioengineering will be extraordinary. References 1. Schwan HP. Organizational development of biomedical engineering. IEEE Eng Med Biol.1991;10:25-29.Google Scholar 2. Fagette P. Tracking the historical development of biomedical engineering: the 1960s and 1970s. IEEE Eng Med Biol Mag.1997;16:164-175.Google Scholar 3. Bronzino JD. The Biomedical Engineering Handbook. 2nd ed. Boca Raton, Fla: CRC Press and IEEE Press; 2000. 4. Hounsfield GN. Nobel award address: computed medical imaging. Med Phys.1980;7:283-290.Google Scholar 5. Cormack AM. Nobel award address: early two-dimensional reconstruction and recent topics stemming from it. Med Phys.1980;7:277-282.Google Scholar 6. Lauterbauer PC. Image formation by induced local interactions. Examples employing nuclear magnetic resonance. 1973. Clin Orthop.1989;(244):3-6.Google Scholar 7. Lardo AC. Real-time magnetic resonance imaging: diagnostic and interventional applications. Pediatr Cardiol.2000;21:80-98.Google Scholar 8. Wells PNT. Physical Principles of Ultrasonic Diagnosis. London, England: Academic Press; 1969:282. 9. Yang GZ, Firmin DN. The birth of the first CT scanner. IEEE Eng Med Biol Mag.2000;19:120-125.Google Scholar 10. Severinghaus JW, Kelleher JF. Recent developments in pulse oximetry. Anesthesiology.1992;76:1018-1038.Google Scholar 11. Spelman FA. The past, present, and future of cochlear prostheses. IEEE Eng Med Biol Mag.1999;18:27-33.Google Scholar 12. Geddes LA, Hamlin R. The first human heart defibrillator. Am J Cardiol.1983;52:403-405.Google Scholar 13. Zollikofer CL, Antonucci F, Stuckmann G, Mattias P, Salomonowitz EK. Historical overview on the development and characteristics of stents and future outlooks. Cardiovasc Intervent Radiol.1992;15:272-278.Google Scholar 14. Yannas IV, Lee E, Orgill DP, Skrabut EM, Murphy GF. Synthesis and characterization of a model extracellular matrix that induces partial regeneration of adult mammalian skin. Proc Natl Acad Sci U S A.1989;86:933-937.Google Scholar 15. Lanza RP, Langer R, Vacanti J. Principles of Tissue Engineering. 2nd ed. San Diego, Calif: Academic Press; 2000. 16. Zhang S, Holmes T, DiPersio M, Hynes RO, Su X, Rich A. Self-complementary oligopeptide matrices support mammalian cell attachment. Biomaterials.1995;16:1385-1393.Google Scholar 17. Kisiday J, Jin M, Hung H, Kurz B, Zhang S, Grodzinsky AJ. Self-assembling Peptide Scaffold for Cartilage Tissue Engineering. Vol 26. San Francisco, Calif: Transactions of the Orthopedic Research Society; 2001. 18. Grodzinsky AJ, Levenston ME, Jin M, Frank EH. Cartilage tissue remodeling in response to mechanical forces. Annu Rev Biomed Eng.2000;2:691-713.Google Scholar 19. Cowin SC, Moss-Salentijn L, Moss ML. Candidates for the mechanosensory system in bone. J Biomech Eng.1991;113:191-197.Google Scholar 20. Davies PF. Flow-mediated endothelial mechanotransduction. Physiol Rev.1995;75:519-560.Google Scholar 21. Young RA. Biomedical discovery with DNA arrays. Cell.2000;102:9-15.Google Scholar 22. Brown CS, Goodwin PC, Tong J, Sorger PK. Image metrics in the statistical analysis of DNA microarray data. Proc Natl Acad Sci U S A.In press.Google Scholar 23. Gygi SP, Rochon Y, Franza RB, Aebersold R. Correlation between protein and mRNA abundance in yeast. Mol Cell Biol.1999;19:1720-1730.Google Scholar 24. Pandey A, Mann M. Proteomics to study genes and genomes. Nature.2000;405:837-846.Google Scholar 25. Lauffenburger DA, Fallon EM, Haugh JM. Scratching the (cell) surface: cytokine engineering for improved ligand/receptor trafficking dynamics. Chem Biol.1998;5:R257-263.Google Scholar 26. Riccobene TA, Omann GM, Linderman JJ. Modeling activation and desensitization of G-protein coupled receptors provides insight into ligand efficacy. J Theor Biol.1999;200:207-222.Google Scholar 27. Griffith LG. Polymeric biomaterials. Acta Materiala.2000;48:263-277.Google Scholar 28. Stokes CL. Biological systems modeling: powerful medicine for biomedical e-R&D. Am Inst Chem Eng J.2000;46:430.Google Scholar
TI - Advances in Biomedical Engineering
JF - JAMA
DO - 10.1001/jama.285.5.556
DA - 2001-02-07
UR - https://www.deepdyve.com/lp/american-medical-association/advances-in-biomedical-engineering-x03lovhreA
SP - 556
EP - 561
VL - 285
IS - 5
DP - DeepDyve
ER -