The Potential of Quantum Computing and Machine Learning To Advance Clinical Research and Change The Practice of Medicine
The Potential of Quantum Computing and Machine Learning To Advance Clinical Research and Change The Practice of Medicine
of machine learning, noting that there is a machine cells without damaging healthy ones. 19 Quantum
learning spectrum which runs from how much a computers would allow rapid and more precise
predictive algorithm is driven by human decisions radiation plans and allow comparisons between all
vs. decisions based on data. A statistical model possible approaches. The end result would be an
requires human decisions about variable definitions ideal radiation dose pointed at the right target and
and their inter-relationships. In machine learning, thus lead to more effective treatment with reduced
there are few assumptions and algorithms are derived side effects. 19,20
by computer programs that evaluate millions of Machine learning and artificial intelligence have
data elements and all their potential directions of been used with traditional computing resources to
effect and interactions. The more an algorithm is interpret CT scans, aid surgical procedures and to
derived from raw data and with less human input, analyze big data to develop predictive models of
the more it fits into machine learning. Machine disease. Gulshan et. al. 21 used machine learning
learning that informs clinical practice in real time and nearly 10,000 retinal images to create an
depends on growing databases containing constantly algorithm that can detect diabetic retinopathy with
updating medical record information and linked to high sensitivity and specificity. The authors note
other sources of data (e.g. wearable technology). To that this is an experimental design with no known
deal with this complexity future machine learning clinical utility. If programs such as the one created
programs will require computational power of by Gulshan and colleagues 21 had the computational
quantum computing to deliver results in real time. power of quantum computing, one could imagine
With more data informing predictive models and running an algorithm in real time to help physicians
quantum computer power, the clinician may be able accurately detect retinopathy, possibly at an earlier
to select a therapy perfectly suited to an individual stage.
patient based on models continuously running Machine learning holds great promise in
and updated to predict treatment response while eliminating most diagnostic errors related to reading
accounting for the numerous patient characteristics imaging. Dilsizian and Siegel 2 describe how artificial
(race, age, gender, comorbidities, co-medications, intelligence can overcome the source of diagnostic
genetic make-up, etc.) which contribute to outcomes. errors in medical imaging. Computers overcome
With the first quantum computing devices human sources of error such as: being stuck on initial
becoming operational in around 2000, Parsons 18 impression, over-reliance on context, drawing quick
speculated that quantum computing applications conclusions based on a previous case and failure to
would generate a new era in medical imaging. It is consider other answers once a conclusion is made.
expected that a quantum MRI machine will generate The volume of material and speed of new
extremely precise imaging allowing visualization research discoveries is over whelming. Even the best
of single molecules. 3 Using artificial intelligence, attempts at providing guideline driven, evidence-
quantum computing can be applied to interpreting based medicine are challenged by the rapid creation
diagnostic images. Not only will image detail be of new evidence. Combining big data with quantum
exponentially improved but the physician can be computing will provide access to the current evidence
aided in interpreting results because efficient machine and enable meaningful use of the electronic data
learning can train a quantum computer to identify continuously generated in the deliver y of care.
abnormal findings with precision better than the Realization of personalized medicine will need to
human eye. 2 draw on analysis of mega-data and bring together
In cancer treatment, quantum computing will measures of physiology, imaging, genomics, wearable
contribute to improved therapies. Computers are technology, screening measures, patient records,
currently used to deal with the thousands of variables environment measures and more. 20
used to develop a radiation plan that targets cancer Quantum computing holds no value to medicine
without the parallel increase in availability of level of access is available, quantum computing will
clinically meaningful data from numerous sources. remain the domain of researchers, not physicians.
Challenges to realizing a healthcare system where
quantum computing uses decades of data from References
1. IBM Watson Health: Enable Effective Care. https://www.ibm.com/watson/
billions of clinical encounters and recommends a health/value-based-care/enable-effective-care/. Accessed April 1, 2018, 2018.
specific medication for a patient presenting with new 2. Dilsizian SE, Siegel EL. Artificial Intelligence in Medicine and Cardiac
onset depression, cancer, diabetes or any condition, Imaging: Harnessing Big Data and Advanced Computing to Provide
Personalized Medical Diagnosis and Treatment. Curr Cardiol Rep. Jan
will be overcome when data can be merged from 2014;16(1).
multiple sources including existing patient data 3. Sullivan MD, Edlund MJ, Fan MY, Devries A, Brennan Braden J, Martin
BC. Risks for possible and probable opioid misuse among recipients of
networks, bio-banks, and wearable health devices, chronic opioid therapy in commercial and medicaid insurance plans: The
etc. TROUP Study. Pain. Aug 2010;150(2):332-339.
Will the future include visits to virtual 4. M.A. N, I.L. C. Quantum Computation and Quantum Information.
Cambridge: Cambridge University Press; 2010.
physicians? Most patients expect more than a pill 5. L.K. G. A fast quantum mechanical algoristhm for database search. STOC:
when they seek care. Particularly when treatment ACM; 1996:212.
6. Shor PW. Polynomial-time algorithms for prime factorization and
involves greater risk as in surger y and when treatment
discrete logarithms on a quantum computer. Siam J Comput. Oct
effectiveness requires patient adherence to complex 1997;26(5):1484-1509.
regimens. Quantum computing does not replace 7. Cleve R, Ekert A, Macchiavello C, Mosca M. Quantum algorithms
revisited. P Roy Soc a-Math Phy. Jan 8 1998;454(1969):339-354.
the doctor-patient relationship, transparency and 8. Devoret MH, Martinis JM. Implementing Qubits with Superconducting
trust required in shared decision making. Nor will Integrated Circuits. Quantum Inf Process. Oct 2004;3(1-5):163-203.
a computer replace the ability of a physician to offer 9. Rigetti C, Gambetta JM, Poletto S, et al. Superconducting qubit in a
waveguide cavity with a coherence time approaching 0.1 ms. Phys Rev B. Sep
encouragement, empathy, social support and most 24 2012;86(10).
importantly providing words of encouragement and 10. Kennedy TA, Colton JS, Butler JE, Linares RC, Doering PJ.
Long coherence times at 300 K for nitrogen-vacancy center spins in
a pat on the back, a handshake, a smile and even a diamond grown by chemical vapor deposition. Appl Phys Lett. Nov 17
hug. We must never underestimate the power of the 2003;83(20):4190-4192.
11. Koehl WF, Buckley BB, Heremans FJ, Calusine G, Awschalom DD.
social contract patients make with physicians and
Room temperature coherent control of defect spin qubits in silicon carbide.
the support of doctors, nurses, staff play in healing. Nature. Nov 3 2011;479(7371):84-U108.
Computer algorithms, whether run on a “traditional 12. Sullivan MD. Who gets high-dose opioid therapy for chronic non-cancer
pain? Pain. Dec 2010;151(3):567-568.
computer” or a future quantum laptop computer, do 13. Bennett CH, Brassard G. Quantum cryptography: Public key distribution
not have feelings and will never be able to provide the and coin tossing. Theor Comput Sci. Dec 4 2014;560:7-11.
14. B. K, C.C.W. L, R. H, et al. Provably secure and practical quantum key
impact of a surgeon’s promise to “get you through distribution over 307 km of optical fibre. Nat Photonics. 2015;9:163-168.
this procedure” or the reinforcing effects of a 15. Pugh CJ, Kaiser S, Bourgoin JP, et al. Airborne demonstration of a
primar y care provider’s encouragement to “walk just quantum key distribution receiver payload. Quantum Sci Technol. Jun 1
2017;2(2).
another 5 minutes” to enhance your health. 16. Yin J, Cao Y, Li YH, et al. Satellite-based entanglement distribution over
1200 kilometers. Science. Jun 16 2017;356(6343):1180-1184.
17. Beam AL, Kohane IS. Big Data and Machine Learning in Health Care.
Summary Jama-J Am Med Assoc. Apr 3 2018;319(13):1317-1318.
We may be at the dawn of a revolution in 18. Parsons DF. Possible Medical and Biomedical Uses of Quantum
computing. At this time we can only point to Computing. Neuroquantology. 2011;9(3):596-600.
19. Pariente P. Quantum computing set to revolutionise the health sector.
experiments and speculate how quantum computing L’Aterlier BNP Paribas; 2015.
could improve imaging, diagnosis, treatment, and 20. A R. Quantum Computing and Health Care. Vol 2018: BMJ Technology
Blog; 2017.
population health. It is yet to be determined if and 21. Gulshan V, Peng L, Coram M, et al. Development and Validation of a
when quantum computers will be accessible for daily Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal
use in medicine and research. We have numerous Fundus Photographs. Jama-J Am Med Assoc. Dec 13 2016;316(22):2402-
2410.
examples of machine learning algorithms and artificial
intelligence that may leverage the power of quantum Disclosure
computing to deliver real time results. Until this None reported. MM