Computer-aided medicine revolution

April 2017

Medical professionals checking resultsDr. Daniel J. Thomas CSci CEng FBCS paints a picture of how augmented reality, artificial intelligence and other cutting edge technologies can save lives and possibly the NHS too.

We each consist of approximately 37.2 trillion cells, 200 different types of cell, 78 organs and 206 bones. True to say, there is nothing more complex on this planet than the human body. Each of our biological systems is vital to sustain us, as is the diagnosis, repair or replacement of these systems when needed.

Digital medicine

The 21st century medical environment is complex. To maintain this, the NHS costs £116 billion per annum (2015), of which 75 per cent of this budget is spent on the cost of treating chronic diseases. It is estimated that healthcare costs will rise to 35 per cent of UK GDP by 2050 unless innovative methods of diagnosis and treatment are found.

Because of this, the government has highlighted that the ‘management of chronic conditions is a necessary and key part of achieving sustainable, effective and efficient health and social care services’.

The integration of information technology in the medical sector offers the evolution in information-focused healthcare, which can potentially lead to a revolution in patient treatment.

Computer-aided medicine

The NHS has truly embraced and integrated information technology into all parts of its service. From patient record data, which are now stored electronically, to the potential use of artificial intelligence (AI) technology such as that of DeepMind envisaged by Google (which is planned to directly be able to analyse huge amounts of patient and operational healthcare data). We are now entering into an exciting dawn and the powerful landscape of computer-aided medicine.

In the most fundamental of forms, using computer systems to sift through huge datasets to establish patterns could offer a vital proving ground before its progression into other vital areas. This would lead not just towards the conventional yet vital applications of managing bed space and medication tracking, but this powerful partnership could evolve to identify patterns in degenerative diseases, detect and track ward infection trends and could help the NHS in minimising outbreaks of disease.

Also among the most useful possibilities is in reading medical images and integrating this information to be used in the vital services. This remains to be the most powerful near-term potential for everyday medicine.

Augmented reality interfaces

Although the complexity of modern-day medicine is now becoming increasingly difficult, can we bring all of this information directly to the hands of the medical professional during everyday life to lend a digital hand?

The first current glimpse into this approach is through the use of augmented reality interfaces (ARI), which remains to be the next frontier in on-demand information lead medicine. This can allow for the different elements in the digital medicine landscape to be integrated together.

The efficient use of data to become interactive 3D information can be directly used towards enhancing medical outcomes. The next stepping stone is on towards integrating augmented reality devices for the purpose of the presentation of real time information.

During the preoperative phase of a procedure, the surgeons will have a mental image of what needs to be carried out. Marking structures of interest on radiographic images that can be superimposed on live video camera images allows a surgeon to simultaneously visualise the surgical site and the overlaid graphic images, creating a so-called semi-immersive environment.

ARI offers a revolutionary and dynamic approach towards aiding the execution of complex surgical procedures and can even assist in medical diagnosis. ARI techniques have been tested and successfully validated in cardiovascular surgical procedures, for example, the goal of improving repairs to the heart mitral valves and replacement of aortic valves more efficiently.

At the current rate of evolution on innovative technologies, augmented reality is pitted to become a critical and useful tool in the operating room. ARI offers the following integrated systems:

Precision multi-axis motion sensing

  • High definition integrative displays
  • Data handling capability
  • Speech recognition

It is important to note that ARI devices include speech recognition technology. This form of non-contact control is vital because it is the most rapid means of accessing information, which does not require any necessary physical contact. Earlier this year the NHS encouraged virtual blood donations with augmented reality outdoor ads via smart phone integration.

In breast cancer surgery, ARI visualisation has been effective in showing superimposed 3D tumour models onto live video images. This has enabled surgeons to see the exact position, in a three-dimensional space, of tumours as if they were visible through the skin.

Although its application is in a preliminary stage, further research is needed to evaluate its long-term clinical impact on patients and surgeons. We have only really scratched the surface of what augmented reality is capable of in the hospital. 

Other variables such as graphics quality, processing power, software durability and dynamic object tracking will become important critical factors for the advancement of augmented reality in the medical world.

Point of care for instant diagnosis

In conventional cases, in order to diagnose disease this usually requires time consuming laboratory tests. Subsequently, there is the requirement for interpretation and subsequently relaying these results back to the medical practitioner; this all takes time.

For example, glucose sensors have revolutionised the way in which diabetes is managed. Now point-of-care technology can be used to diagnose conditions, such as - cardiovascular disease, cancer and meningitis to a high degree of sensitivity and selectivity.

An antibody-based biosensor is a device which can detect and quantify target molecules. These systems allow for instant diagnosis to be made through the process of selective sensing of disease biomarker triggers. This highly sensitive means of detecting disease is achieved through monitoring individual antigen–antibody interaction.

These devices are fabricated by attaching a biological receptor material, for instance an antibody onto the surface of a suitable transducer. This converts the biochemical signal when a complementary disease biomarker binds to this surface into quantifiable electronic signals.

To ensure that these devices are able to detect the presence of disease down to the 0.1pg/ml concentration in a sample then high aspect ratio nanowire-based detection strategies provide promising new routes. This allows for the rapid bioanalysis and detection of even serious life threatening conditions in the community, surgery or even at home.

Artificial intelligence

AI has continued to progress towards practical applications in healthcare. The next stage of evolution is on applying machine learning to the fields of disease diagnosis and subsequent treatment.

As AI starts to become present in everyday life, it could be an effective means towards revolutionising the healthcare system. It has already proven useful in various scientific fields and in the elemental diagnosis of cancer, autism and mental health disorders.

It is this data-driven medicine approach that has the potential to boost the speed and accuracy of disease diagnosis. With development this opens up a new paradigm, which is the ability to make treatments patient-specific.

Early integration of AI is now being applied to searching through previous cancer research results to help radiologists in utilising machine learning to monitor the progression of tumours. Subsequently this is evolving towards the development of algorithms to assist in planning treatment options for different types of cancer.

These current artificial-intelligence-driven initiatives are aimed at assisting medical experts in research, diagnosis and progress in looking for potential cures for different cancer types but with existing treatments.

Future perspective

These new information-led technologies for medical assistance offer significant potential. The amalgamation of medical data to form real time information can provide vital knowledge to those who need it, when they need it. Computer aided systems have already revolutionised financial institutions, large international manufacturers and the construction sector.

Although modern healthcare is a multifaceted complex field, there is significant influence that information-led technologies can lead towards. The integration of key technologies and their incorporation into a range of healthcare application has potential to revolutionise and enhance both diagnosis and treatment for a new era.



Doctors with Data (infographic)

Comments (7)

Leave Comment
  • 1
    Sabine McNeill wrote on 11th May 2017

    It would be nice if BCS could help get my innovation of re-visualising images as the basis for new insights and investigations into the world of medicine!

    See for example.

    Report Comment

  • 2
    Phil wrote on 11th May 2017

    When I read this article, based upon my recent experience as a patient in the NHS, I said to myself " this guy must be having a laugh!!".

    I admire the author's enthusiasm and I am sure all these things he talks about are possible. Just because they are possible does not mean they will happen. My recent experience shows how far away, much of the NHS is, from what the article proposes.

    Let me give my personal experience.

    In 2013 I had an abscess on my tooth. Because of the position of the abscess and other symptoms my GP got involved as well as the dentist. I was sent for a CT scan at a local private hospital but on the NHS. The CT scan showed my teeth are particularly close to my sinuses and the ENT consultant who I saw said " whoever removes the tooth would have to be careful not to make a hole from the mouth into the sinuses".

    Accordingly my Dentist referred me to the local Dental Maxillo-Facial Surgery team as if there were problems they could fix it on the spot. My tooth was removed by them.

    In November 2016 I had a similar problem to that in 2013 - the same tooth the other side of my mouth. My usual dentist had moved from the practice and I saw someone else. I had a swelling on my gum and X-rays showed a failed root canal treatment and I was referred to a endodontic specialist, with a view to a possible re-doing of the Root Canal. This was a private referral as the NHS does not re-do root canal treatments. I was referred mid December and nothing happened for 8 weeks. I phoned my dentist twice and was assured I had been referred by email. Eventually I contacted the place to which I had been referred and they had never heard of me, or any referral from my dentist. My dentist then faxed my referral to the endodontist and I got an appointment in a 4 weeks. Clearly the communication system between the 2 parties is not robust. The conclusion by the endodontist was that the tooth could not be saved and should be removed.

    I went back to my dentist and mentioned the close proximity of my teeth to my sinuses - assuming that my CT scan etc. would be available in the NHS - neither my doctor nor the dentist had access to it.

    I telephoned the Hospital where I had the CT scan done and they sent me the CT scan - cost £15 from which I extracted the relevant information on my computer as a jpeg file. I emailed this to my dentist who then referred me to the Maxillo-Facial Surgery department - as in 2013.

    I had an appointment at this department earlier this week. The consultant had no details of my 'encounter' with the endodontist nor of my CT scan. He could not access details of previous visit to the department in 2013. He could not access details other visits I made to other departments in the hospital. He had a very large paper file comprising lots of hand written notes - to which the consultant added even more. I was sent for an x-ray - which would not have been needed if the CT scan had been available.

    I was told they will remove my dodgy tooth in 6 to 8 weeks - but it may be even longer because a new IT system is being introduced and that is bound to delay things. Perhaps I shall get rid of this tooth by the end of July if I am lucky. That will make it 8 months from start to finish. I am so glad the condition isn't life threatening!

    It seems to me that the NHS needs to get the simple things in IT sorted out - like access to patient records, simple messaging etc. before it starts anything fancier.

    Report Comment

  • 3
    Dr M J C Brown wrote on 11th May 2017

    The writer of this article, Dr Thomas, apparently without a medical degree that the editor feels worth displaying, makes bold sweeping assertions without any evidence whatsoever to support those assertions. Even in my short time a a doctor, I saw the rise, abuse and fall of several technologies despite their being apparently most patently applicable and beneficial. Controlled clinical trials are needed to properly assess any expensive or medical intervention. And just in case you wonder how can one perform controlled clinical trials on computer based system, I would point out that in 1969 (nineteen sixty-nine) Bulpitt, Coles and Byles reported on a controlled clinical of computer held hypertension records which had been performed at King's College Hospital ( yes I know doctors are famously so slow to adopt new technology). There are plenty of indications that computerisation of medicine will have and hashad adverse on medical practice. As for 'big data' the Shipman case well illustrate what may happen. In the case of Dr Shipman, other general practitioners were the FIRST to draw attention to the unusual pattern of deaths in his practice. The local police force trawled through the computer held data concerning deaths and found nothing to warrant further investigation and dismissed the general practitioners' concerns. Only when the general practitioners renewed their efforts to bring their concerns to the attention of higher authorities, was the same computer held data trawled through and this time found to support the general practitioners' concerns. How many lives would have been saved (if any), had the police acted on the knowledge of the general practitioners rather than relying on computer held data. One could go on .... and on, but I will not.
    Dr M J C Brown MB,BS. M.B.C.S

    Report Comment

  • 4
    Matt Durcan wrote on 11th May 2017

    In fairness to the author, this piece is intended as a taster of what might be achieved, rather than a state of the NHS summary.

    1. Without great consideration, here's a link that might be of interest to Sabine:

    2. Tales like this where the patient takes ownership, control and technical responsibility for filling in the gaps in the system are saddening and maddening - not least because the majority of citizens do not have the resources to do so. I wonder quite frequently whether these inefficiencies are actually exploited to extend waiting times and control load, but delay costs money - admin effort, not to mention increased cost of duplicated effort, deferred treatment and patient distress. This is where a great deal of effort is focused today and I hope we can all agree on its general utility. But the NHS is a victim of its own success and simply enormous scale.

    3. I wonder whether after "a long time as a doctor" the 3rd contributor might feel that many of the tasks they undertake could be automated (get rid of the easy automatic bits so that skills can be concentrated) so that they can focus their training, skills and experience on the cases that really require medical judgement - rather than spreading thinly across all patient presentations. Of course how this is done requires Clinician input and validation - and I do hope they will volunteer for appropriate trials.... otherwise the IT team will relegate you to a role as Dr House!

    Report Comment

  • 5
    Keith wrote on 2nd Jun 2017

    Some interesting perspectives, especially the significant burden of long term and chronic disease treatment.
    From my own experience, as a technologist working alongside clinical teams (in a non-technical role), it is my experience that there is *nothing* in a clinical role that can be automated. No doctor acts on auto-pilot. Every consult, every decision has to be thought through, because every patient, no matter how routine, is different. Even the most mundane note taking requires care and remains a unique process for every patient.
    The other point to note is the huge difference between primary care and specialist treatment.
    The emphasis on using technology, AI etc, in specialist treatment is exciting and has huge potential. But in primary care there are far fewer opportunities, and much less to be gained.
    With regard to the issue of comms and sharing data etc., the healthcare system is like any other industry, it suffers from the involvement of human beings!

    Report Comment

  • 6
    Dr. John Evans wrote on 19th Jun 2017

    An excellent article, very thorough and I really enjoyed reading it. As a surgeon on forty years and later a general practitioner, I am all for the adoption of new technology. The only obstacles that I see towards the adoption of AI and POC technology more widely is by jaded medics and some patients. The cynical point of views by these individuals have over the adoption of new technology is akin to the prehistoric humans who would have discovered fire and then made the comment that it is good but they don’t like it because when I put my hand into it, then it catches on fire. As Phil and Dr M J C Brown suggest and confirm, past experience should not be a prelude towards future progression – otherwise nothing would ever change.

    Report Comment

  • 7
    Nicky James wrote on 20th Jun 2017

    Interesting that Dr M J C Brown also makes sweeping assertions without any evidence whatsoever to support those assertions. As he/she suggests that Dr Thomas, apparently without a medical degree. How do you know Dr Thomas doesn't have a medical degree - or are you just making an assertion?

    Report Comment

Post a comment