Article Text
Abstract
There is increasing interest in the utility of digital pathology in the diagnostic setting. Successful transition requires guidance and training, but additionally an understanding of opinions and attitudes of histopathologists to ensure that potential barriers are addressed. Histopathology trainees as a group are likely to be at the forefront of this revolution, and have specific and as yet largely neglected training needs in this context. We designed an online survey for trainees within our region to capture their opinions and attitudes to digital pathology in the diagnostic setting, and to assess their perceived training needs. This survey indicates overall that these trainees have similar aspirations with regard to the predicted utility of digital pathology and the challenges faced as have been recognised among consultant histopathologists. While their training needs are also largely similar, there are specific additional considerations based around training in multiple centres with varying exposure to digital pathology.
- digital pathology
- training
- histopathology trainees
Statistics from Altmetric.com
Introduction
There is increasing interest within the pathology community in the utility of whole slide imaging (WSI) as the basis for digital histopathology.1–3 WSI technology has been used in an educational context for many years1 4 5 and UK histopathologists are familiar with WSI in mandatory external quality assessment (EQA) of their work. Although WSI is currently in limited use within the clinical diagnostic setting for surgical pathology, there is rapid expansion of this application into diagnostic practice in the UK. As yet no National Health Service (NHS) histopathology laboratories are fully digital however.
We are rolling out the use of digital pathology in diagnostic practice in our tertiary referral centre, with the intention of running a fully digital diagnostic service in an NHS setting. A multiprofessional Digital Pathology Steering Group was established to oversee the pilot study and the ongoing transition phase, which comprises consultant histopathologists and senior laboratory management. This group has identified the histopathology trainees as a group who have specific and as yet largely neglected training needs in the context of diagnostic digital pathology. In this context, the group wished to capture the opinions and attitudes of these trainees to digital pathology and to gain a better idea of their perceived training needs in order that these could be addressed.
Methods
A survey was circulated via the online SurveyMonkey survey tool (www.surveymonkey.com) to 19 trainee histopathologists in their second year of training and above, based in four regional hospitals; the index tertiary referral centre with access to digital pathology, and three district general hospitals currently without access. The group comprised 16 Health Education England specialty trainees and 3 specialty doctors who participate in training activities within region. First year specialty trainees were excluded; these trainees have a teaching programme which differs from the more senior trainees who attend regional training sessions and access digital slides for teaching in this setting.
The aim was to gather trainees’ opinions with regard to the implementation of digital pathology within the diagnostic service, to assess current experience using digital pathology, and to explore their perception of future training needs. The survey results would be used to inform the development of a training programme in digital pathology within the region, and, through better understanding their perceptions and opinions, potentially improve the success of such training.
The survey was divided into nine ‘sets’ of questions, and in total there were 50 individual questions (online supplementary table). There were general questions centred around current experience of digital pathology and perceived training needs, perception of value of digital pathology, readiness for transition and areas of potential benefit and challenge in the use of digital pathology in the diagnostic setting. There were also four sets of questions focused on areas identified previously as ‘benefits’ of adopting digital pathology6—patient safety related, workflow related, workforce related, service quality related. Free-text comments were invited and encouraged.
Supplemental material
The survey was not intended to explore digital pathology within the research/artificial intelligence setting. The trainees were asked to consider the answers as may be applied to future practise as a consultant pathologist as well as at their current stage of training.
Results
Sixteen trainees responded to the survey (15 specialty trainees and 1 specialty doctor, 84% response rate), and all answered in full.
Relating to current experience of digital pathology (figure 1), none of the trainees have to date had experience in reporting live clinical cases on the digital platform. A total of 15/16 have seen teaching cases on the digital platform and 4/16 have seen EQA cases digitally. A total of 10/16 were aware of the Royal College of Pathologists (RCPath) Guidance on the implementation of digital pathology.7 Only 3/16 agreed that currently they would feel confident in reviewing cases digitally in an examination setting.
The majority of trainees (15/16) agreed that digital pathology in diagnostic practice is a positive step for the profession, and 12/16 thought that they would likely be routinely reporting cases digitally as a consultant pathologist, although 9/16 felt that they were unprepared for the transition.
Benefits of adopting digital pathology
Regarding patient safety, 9/16 agreed that digital pathology would potentially reduce the risk of patient/slide misidentification, and 13/16 agreed it would potentially reduce the risk of tissue/slide loss or damage. Free-text comments were similar to those cited in previous publications6 and included improved access to specialist opinions, ease of multidisciplinary team (MDT) meeting preparation, faster retrieval of archived cases, and ease of collection of cases for teaching.
Considering workflow, only 4/16 felt that digital pathology had the potential to improve turnaround times. Most agreed it would improve access to second opinions and that there would likely be an increase in referral cases into the department (table 1).
Free-text comments suggested
Potential for turnaround times to lengthen during the transition phase.
Workflow impeding factors including workload, meetings, teaching might offset reduction in time to getting cases to a pathologist.
Additional workflow steps introduced, such as slide scanning, might impact turnaround.
Potential for positive impact on MDT meeting preparation time and time for preparing teaching sessions.
There was mixed opinion as to the potential for digital pathology to provide a solution to shortages in consultant workforce (4/16 agreed it might). A total of 15/16 felt that digital pathology offered the opportunity for improved collaboration between departments within the network, and 12/16 felt that it may result in improved collaboration within the department between teams and within teams. One trainee commented that improved collaborations were dependent on the individuals involved and more so than available technology. Improvements in the ability to work from home were identified as a potential benefit to part-time pathologists who may be able to offer additional hours.
Potential improvements in relation to the quality of clinical service focused on MDT working and teaching (figure 2). A total of 12/16 agreed that showing digital slides at MDT meetings might improve the understanding of the MDT members of the pathology being discussed. Free-text comments in relation to this suggested that time available was the limiting factor in whether pathology slides are shown, digitally or on glass. Showing annotated slides was considered potentially beneficial, as was access to archived cases during an MDT for comparison.
The trainees were overwhelmingly positive regarding potential benefits in relation to teaching using digital slides (figure 2). Accessing scanned slides from a clinical platform was considered to be beneficial when compared with the current system whereby selected teaching slides need to be scanned specifically for this purpose and then uploaded onto a separate cloud-based viewing platform, which is time-consuming and labour intensive by comparison. Comments were made around the need to ensure secure storage of images, and around the use of images and data within the boundaries of the legal system and the need for improved guidance on this issue at a potentially national level.
General considerations
This section focused on more general considerations around the transition to digital pathology (figure 3), such as the validation and transition process, risk of error, and technological aspects for remote working. There were no major areas of concern identified, and in particular the level of concern regarding the actual process of transition and the validation process was not high.
Trainees were asked to identify within free text comments, areas of potential diagnostic challenge on the digital platform and areas which may be potentially easier. The results are shown in table 2.
Training needs
Proposed topics for training were presented and trainees were asked to indicate whether these topics were of potential benefit (table 3). All suggested topics were identified as being of potential benefit to almost all trainees.
Additionally, free-text comments suggested that the trainees would like:
Regular access to cases digitally to facilitate familiarity with the platform and digital diagnosis.
A broad package of training to include training on areas known to be a diagnostic challenge.
A means to demonstrate their digital pathology training credentials for their portfolio/CV.
Discussion
The transition to the use of WSI (digital pathology) in the diagnostic setting in the UK is gaining momentum with several NHS histopathology laboratories already on the road to implementing a fully digital service. With challenges faced in the histopathology workforce, investment in the digital pathology infrastructure has been recommended by Cancer Research UK to ‘future-proof’ the specialty.8 With this in mind, the training of histopathologists to work in the diagnostic setting on digital platforms is paramount. The RCPath ‘Best practice recommendations for implementing digital pathology’7 provides guidance on the implementation of digital pathology in a diagnostic setting, including an overview of a pathologist-led self-validation process. This goes some way to advise and inform pathologists embarking on this journey, however, the needs of histopathology trainees are not covered in this document and at the current time no formal guidance exists. Similarly the College of American Pathologists guidance document9 states that ‘Although it is important that users be trained to use this technology, the personnel required for WSI and how they should be trained are outside of the scope of this document’. There is minimal published literature on the topic of training programmes for histopathologists in digital pathology in the diagnostic setting, which includes a single report on a digital pathology summer school held in Italy for which a syllabus on training was proposed.10
In our region, we have used digital slides for regional teaching for over 5 years, and consequently, our histopathology trainees have experience in viewing and interpreting WSI in an educational setting. Furthermore, we have for several years held annual study days on digital pathology for the trainees, covering diagnostic and research applications of this platform. On this background, this survey was designed to gauge the opinions and attitudes of our trainees around the adoption of diagnostic digital pathology, and specifically to explore perceptions of their training needs.
The survey indicates overall that these trainees have similar aspirations with regard to the predicted utility of digital pathology and the challenges faced as have been reported previously.1 6 Based on our experience as a centre transitioning to diagnostic digital pathology, we propose that training needs to incorporate a broad range of topics not limited to validation/training on the interpretation of WSI, and should include topics such as technological aspects of digital pathology, laboratory workflow, governance considerations, and even health and safety—these aspects of training are not different to those of consultant histopathologists. The survey results indicate that the trainees agree these topics should incorporated within any proposed training package. Ideally this training should incorporate real-life experience as far as possible, and will require a multiprofessional approach. While trainees require exposure to WSI on a diagnostic platform in-line with the consultant experience, their needs in this context are different given that their training will be based across centres with varying exposure to diagnostic digital pathology; there needs to be an awareness that at the current time they are unlikely to become ‘fully digital’ in terms of their diagnostic workload as trainees or even as consultants. Any training programme, therefore, needs to consider and maintain this context, and trainees and their trainers will need to be mindful that (currently) they will need to maintain competence with traditional microscopy to accommodate the current setup of training within multiple centres with differing facilities, and the current FRCPath examination setup (part 2 specifically). Consultant pathologists within a centre transitioning to diagnostic digital pathology will similarly likely need to maintain competence on both digital and traditional platforms until all aspects of diagnostic work (surgical pathology, cytopathology, referral and private practice) are digital. An awareness of artificial intelligence applications within this setting is also desirable, with such technologies benefiting from the contributions of histopathologists to their development and implementation.
On this basis, we have proposed a local digital pathology training programme for our trainees to incorporate the above features, and which will facilitate their understanding around more peripheral details to be considered during the implementation of digital pathology within a diagnostic setting in the NHS (online supplementary file). It will be important to report on and to evaluate this training in due course, with adaption of the programme in-line with any future guidance from the RCPath.
Supplemental material
Conclusion
The opinions and aspirations of trainee histopathologists with regard to digital pathology are similar to those of more experienced histopathologists, and need to be considered during the period of adoption of this technology in the diagnostic setting. While their training needs similarly overlap with those of established consultants, there are additional specific considerations around maintaining competency and confidence with traditional microscopy.
Acknowledgments
With thanks to the histopathology trainees within region for completing the survey.
Footnotes
Handling editor Dhirendra Govender.
Contributors LB confirmed that all authors contributed to this manuscript. Conceptualisation: LB. Acquisition of data: LB, LW and NM. Analysis and interpretation of data: LB. Drafting of the manuscript: LB. Critical revision of the manuscript: LB, RC, JR, LW, NM and CV. Supervision: CV.
Funding This work was supported by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre (BRC).
Disclaimer The views expressed are those of the authors and not necessarily those of the National Health Service (NHS), the NIHR or the Department of Health. CV, RC, LB and JR are part of the PathLAKE digital pathology consortium (JR and CV are PIs). These new Centres are supported by a £50m investment from the Data to Early Diagnosis and Precision Medicine strand of the government’s Industrial Strategy Challenge Fund, managed and delivered by UK Research and Innovation (UKRI).
Competing interests We declare no conflicts of interest
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.