Introduction
Despite us being portrayed as an unsociable specialty, I don’t think many people would disagree that some sort of collaboration is important in our work as pathologists. Collaboration with clinicians is a vital and rewarding part of our jobs, and collaboration within our own departments is essential to a smooth-running service. In histopathology we are akin to a factory production line, with our clinical colleagues providing the specimens, porters and reception staff responsible for getting them into our departments with lots of administrative and laboratory processes in between specimen arrival and report sign out.
Most of us working in the NHS and beyond now work as part of a regional or national collaboration with other hospitals and remote specialist services. From my experience as a regional upper GI pathology specialist lead, poor national infrastructure created many of the problems affecting the efficiency of the regional service.
In 2021 I was asked by Raji Ganesan in Birmingham to help her write an article about adaptation in an innovation during COVID-19 (in the journal of Diagnostic Histopathology) with a few of her colleagues in the Midlands. She said she had been impressed with my IT skills during our previous remote interactions and case discussions and encouraged me to be the primary author and primarily talk about my own pet topic of collaborative digital pathology (DP).
Rather than emailing numerous copies of the article back and forward to each of the five authors I suggested we “dynamically collaborate” by editing the same document shared in a group chat using the (at that time) newly established NHS teams software. I am often surprised to find that not every NHS employee is aware they have access to this facility , which includes their own OneDrive cloud storage through their secure nhs.net login. Any file can be viewed, edited and reviewed by numerous collaborators dynamically if you are all part of the network. If you are logged on at the same time, it is even possible to see your collaborator’s cursor with their initials moving around editing the text. We use this Microsoft facility in one of our regional MDTs, with all the agendas and outcomes viewable and editable in Microsoft’s secure NHS cloud. Instead of distributing loads of attached files each week which quickly clog up the storage space of everyone’s account, we receive a link which opens the single live document on the cloud. Using links means you know you are always looking at the most up to date list, especially if a patient is added or removed from the discussion a few hours before the meeting. If you make a change to the outcomes, the edits can be easily seen, time stamped with the editor’s name marked when viewed in the collaborator mode. It is also a much more secure method of sharing clinical documents than attached documents, since the latter can be more easily be accidentally shared to (and viewed by) a non-NHS employee. Alternatively, if a link is shared (even to an NHS employee who is not part of the MDT team) the document will not open.
The Elephant in the Room. Is NHS digital pathology achieving all the benefits predicted?
DP was predicted to be the immediate savior of the fragmented NHS organisational structure, but unfortunately physical slides are still regularly posted around NHS trusts, even between those who are fully digital, for several reasons. Firstly, unless you are lucky enough to be part of NPIC (National Pathology Imaging Cooperative) then every NHS Trust works as a separate entity particularly their IT security departments, hence when a DP system is installed, even if it is cloud hosted it will usually have an additional layer of security in the form of a VPN (virtual private network) for their own employees only. Although NHS pathologists will usually be able to use the technology to work from home (if they are using the NHS laptop with VPN), they will be unable to share the slides to someone outside of their own organisation. Secondly, most DP systems lack the universal DICOM format used in radiology that carries demographics as well as imaging data; DP is moving towards DICOM but currently all DP vendors use their own proprietary file formats. Thirdly, our files are much larger than those in radiology, so although they can be easily streamed using DP software, they can’t easily be moved around separately as files. Sadly, that means that the digitised urgent case you wanted to send for an expert opinion to a colleague in a different hospital will often have to be sent as glass slides instead. These slides are then often scanned again at the remote trust or private organisation at additional cost and time burden. Likewise, I know it to be the case that even fully digitised trusts are (for the same reasons) outsourcing glass slides to private “backlogs” companies to be reported, with the companies scanning the slides again and distributing them to NHS consultants to report in their free time! Scanning and storing images multiple times seems to me to be a ridiculous waste of money and expensive resources when such a huge investment of money and time has already been invested in NHS trusts to try and prevent and cut down on outsourcing and remote second opinions.
Expensive fragmented procurements with glacial timescales, ring fenced IT infrastructure with exaggerated security concerns, VPN-limited architecture and lack of slide sharing interoperability are major barriers to realising the full potential of DP (and AI) in the NHS. Darren Treanor and his NPIC team in Leeds are at the forefront of what I think is an important and optimistic alternative to the otherwise siloed DP projects springing up all over the NHS. NPIC are trying to cooperate nationally in a more efficient way, with DP and AI now spreading across the North of England, benefiting from economies of scale. If all NHS trusts could join their network, then we would be some way to alleviating the problems we currently face. I fear however that especially with the dissolution of NHS England that there may be insufficient prioritisation of national pathology projects by the government to ensure it is rolled out everywhere in my lifetime at least. A now retired colleague of mine once said to me (long before COVID19!) when I was in the depths of despair that we would never get DP “don’t worry Tim the money always shows up at time of crisis or scandal” and he has of course been proven right (about the crises anyway!). We know that as soon as cancer turnaround times start to soar in our own trusts, money is immediately offered up for outsourcing. Also, seemingly overnight during the pandemic along with all the other NHS spending, Teams and OneDrive became available and secure collaboration became possible in more ways than just telephone or email. Prior to this, even teleconferencing systems in our southwest hospitals did not work outside each individual hospitals (the same situation many are in with their DP systems)! So, I had a thought some time ago that it must surely be possible to collaborate with using DP nationally if the system were hosted in a certain way.
What is the alternative to procuring Digital Pathology equipment for individual NHS trusts?
In their recently published white paper “Accelerating NHS Digital Pathology Adoption”, private-public partnership models were proposed. This is where NHS trusts jointly commission a DP provider (or consortium) to act as a regional or national hosting and integration partner. It seems to me this could solve many of the current issues. Here’s how such a method could perhaps work:
1. Establish Regional Federated Digital Pathology Hubs
NHS trusts partner with a private company to create a shared, cloud-based digital pathology infrastructure. The provider would host, manage, and secure digital slide archives, integrating with local LIS/LIMS systems. The local LIMS would then continue to send reports back to the electronic patient record as they do at present.
2. Use Vendor-Neutral, Interoperable Cloud Platforms
Implement a vendor-neutral archive (VNA) approach to store digital slides in DICOM format (or DICOM-compatible wrappers). This avoids vendor lock-in might allow AI algorithms, slide viewers, and reporting tools to interoperate across sites, rather than certain tests or technological advances being a “postcode lottery” or only being available at well-staffed university hospitals.
3. Centralised Identity and Access Management
The NHS Single Sign-On method I have already outlined could allow pathologists to access slides securely across organisations without being tied to local VPNs. Access controls could be set per case, per MDT, or per trust policy, not per firewall. This security verification method would require everyone to move to the national NHS.net system (unfortunately many trusts seem to be holding back from doing this which is already hindering collaboration possibilities amongst other national NHS colleagues).
4. Outsourced Scanning, QC, and Data Management
The private provider could handle slide scanning, quality control, and digital archiving, reducing the operational burden on NHS lab staff. Most NHS trusts currently store their digital slides in their own local or privately procured archives rather than using much cheaper big data services like Microsoft Azure, Amazon Web Services even though the former is already used for teams and NHS mail!
5. Flexible Commercial Model with Shared Savings
The company could charge per case or offer a subscription model with discounts based on volume and number of participating trusts. If the system were to be rolled out nationally the economies of scale and benefits to patients with access to specialists anywhere in the network would generate considerable savings.
6. National Specialist Reporting Hubs on work from home days
Private companies could separately employ pathologists (e.g. for one or two PAs a week) to either trawl through national work which is exceeding turnaround times or collaborate with other specialty experts in their chosen subject which may have a national shortage (e.g. Sarcoma or lymphoma). I helped set up a “dynamic digital pull” system for one of the companies which can be set up to allow any cases getting near breaching their waiting times to show up as red in the pull list. The details of the cases (apart from the specialty) are hidden from the pulling pathologist to prevent people preferentially “cherry-picking” the easiest cases. A maximum of four cases can be pulled by each pathologist and if the cases aren’t either reported, referred for second opinion or sent for additional lab work, they go back into the pull list. Since the system has been implemented all the cases in pull disappear almost immediately, even when hundreds of cases are put in the pull pile at the weekends. The reason for this is obvious; it is much quicker for 100 pathologists to report one case each than 1 pathologist to report 100!
Partnering with private companies is often viewed with fear and trepidation by NHS employees, but it is often more cost effective, already happens everywhere in the NHS. A company offering up an entire end-to-end digital pathology service is likely to be more effective in setting up and sustaining the digital service. In contrast current NHS procurements often result in a system where equipment and software provided by multiple companies is dropped off with the assumption that busy NHS staff will be able to operate and maintain it. The technology is usually funded with “left over capital” at the end of the financial year, but subscriptions are a much more cost-effective way of managing digital storage (for example AWS charges fractions of pennies to store large files when they are intelligently moved between hot and cold storage). In my (NHS) experience, the time hardware (and LIMS software) is up and running it is often already out of date but staff must wait for the next business case for the chance of an update. The thought of going back to procurement every few years and making vital managerial, technical and clinical staff sit through endless meetings discussing what is essentially vital equipment and IT infrastructure to help our patients fills me with dread! There must be a logical alternative route.
Conclusion
The case for national collaboration in digital pathology is no longer just aspirational, it’s essential. We already have the tools, the talent, and the urgency. What we lack is the infrastructure and the will to scale innovation beyond individual trusts or local cancer networks. By adopting a joined-up, cloud-based model with universal standards and smart private partnerships, we could transform DP from a siloed local tool into a seamless national service.
I truly believe pathology is a critical pillar of modern medicine, and it is time to stop tinkering around the edges and start building the system our patients, and our profession deserve.
Relevant documents
https://npic.ac.uk/deployment/success
https://sourcebioscience.com/wp-content/uploads/Accelerating-NHS-Digital-Pathology-Adoption.pdf
Dr Tim Bracey RCHT
Digipath conference presentation Edinburgh 2025
Leave a Reply