RPS 1305 - Imaging informatics in Europe and beyond

Purpose:

To describe the national trends in diagnostic imaging used in Spain from 2010 to 2017 in public and private healthcare.

Data was obtained from National Hospitals Statistics, an annual report published on the Spanish Health Ministry webpage (https://www.mscbs.gob.es/estadEstudios/estadisticas/estHospiInternado/inforAnual/homeESCRI.htm). Included variables were annual examinations rate per 1,000 inhabitants, annual scanners/rooms per million inhabitants, and a ratio between the number of examinations per scanner/room. Included modalities were CT, MR, PET-CT, SPECT, digital angiography, mammography, conventional radiography, and scintigraphy.

For cross-sectional imaging, annual national examinations rates per 1,000 inhabitants increased 114.8% for PET-CT (1.62 to 3.48), 51.4% for MR (47.58 to 72.04), 42.5% for SPECT (2.40 to 3.42), and 33.4% for CT (83.13 to 110.92). In 2017, public healthcare accounted for 81% of PET/CT examinations, 61% of MR studies, 87% of SPECT examinations, and 83% of CT.

Conventional radiography and mammography also increased their examination rates by 9.9% and 38.1%, respectively. Conversely, digital angiography and scintigraphy decreased by 1.8% and 12.5%. All modalities except conventional radiography and scintigraphy increased the number of scanners/rooms per million inhabitants.

All diagnostic imaging modalities significantly increased the number of examinations per 1,000 inhabitants in Spain between 2010 and 2017, except digital angiography and scintigraphy.

This is an observational study for a period of 8 years. Data should be used carefully due to amount of aggregation.

n/a

No funding was received for this work.

Purpose:

To evaluate general knowledge and concerns about trends on imaging informatics among radiologists currently working in Spain (both residents and attending physicians).

Type of study: a cross-sectional survey.

Inclusion criteria: a radiologist currently working in Spain (public and private).

Survey preparation: a survey comprising a total of 20 questions was confectioned, inquiring about technical issues and ethical and professional perspectives.

Outcome measures: observational statistical analysis, mainly of frequencies, was carried out.

Most of the 223 surveyed radiologists [52 (23.32%) residents and 171 (76.68%) attending radiologists (AR)] work in National Health Services, at least partly. The most heard-of technologies were machine learning (72.55%) and artificial intelligence (67.26%), with the latter (22.59%) and algorithms (18.39%) being the most used technologies among the surveyees.

Although most of the main commercial technologies are recognised by Spanish radiologists, there is a lack of knowledge of the underlying methods used. There is a will to learn about these topics, although there is no time during residency years. The main concern is patient safety, which highlights the ethical implications of the implementation of such technologies.

The main limitation is the selection bias due to convenience sample. A larger sample would allow to draw better conclusions. A higher participation, particularly among residents, was desired.

n/a

No funding was received for this work.

Purpose:

To present a computer system to assess the justification in radiology that allows us to evaluate the requested examinations, the justification of exposures being applied and determining whether patients fit the recommended criteria for the procedure, and to evaluate its use in a radiology service in the last 6 months

We have a tool in the electronic history that allows the evaluation of radiological requests before they are cited. When the radiologist considers the test is indicated, they proceed to its validation, and then it can be cited. Otherwise, they deny the indication and the test is cancelled or changed to another imaging test more appropriate to clinical suspicion. Any change in the request is communicated to the petitioning physician.

Of the 84,342 requests made in the last semester of 2018, we included in this validation system all the CT, MR, Doppler, biopsy, and puncture tests, which represent 10,342 requests (12.2%). 10,044 requests were considered indicated but were not indicated 249 (0.29%). We analyse the results by radiological modality, service, petitioning physician, and diagnostic suspicion. We evaluated those tests that, not being indicated, have been modified to other tests more adjusted to the clinical suspicion.

Justification of radiographic examinations is the practice of evaluating requested radiological examinations to assess for clinical merit and appropriateness based on clinical notes and patient information.

We believe that our validation process has been useful in the radiological management of the patient, in reducing the radiation received, and in avoiding unexpected findings that divert the main diagnosis in unjustified tests.

Radiologist who actively participate in the decision‐making process of justification of an examination will ultimately contribute towards improved patient care and management.

n/a

n/a

No funding was received for this work.

Purpose:

Social media use has increased substantially over the last decade, particularly in medical professional fields. We intended to characterise current social media use for professional purposes in the international radiology community.

A web-based survey was accessible from April to July 2019 and included questions on demographics, professional background, and social media use. Multivariable logistic regression was used to assess independent predictors for professional social media use.

Participants (n=1,041) were working in 54 different countries, mostly Europe (n=867, 83%). Slightly more than half (n=553, 53%) indicated the use of social media for professional purposes. Among participants who used social media, LinkedIn was most frequently mentioned (n=360, 65%), followed by Facebook (n=199, 36%), Twitter (n=115, 21%), and Instagram (n=99, 18%). Professional social media use was inversely associated with age (OR 0.98 (0.97-0.99), p<0.001), and positively associated with advanced scientific training (PhD/research fellowship; OR 1.65 (1.24-2.18), p<0.001) after adjusting for gender and region. There were no significant differences in professional social media usage between residents, fellows, and radiologists; academic and non-academic participants or gender. Male gender was a predictor for Twitter use (p<0.05), female gender for Instagram use (p<0.05). Working in Europe was a predictor for Twitter and Instagram usage (p<0.001). Advanced scientific training was a predictor for LinkedIn and Twitter (p<0.001). There were no predictors for Facebook.

Our international survey indicates radiologists and residents with advanced scientific training more frequently use social media for professional purposes, especially LinkedIn and Twitter. European radiology professionals more frequently use Instagram or Twitter for professional purposes.

"Social media use for professional purposes" is subjective as indicated by the participants.

n/a

No funding was received for this work.

Purpose:

Peer review is seldom reported in France and double reading has been recommended in teleradiology. We therefore report our peer review experience among a community of teleradiologists.

Peer review was performed anonymously and on a voluntary basis. Over one year, randomised CT and MRI cases were reviewed within our RIS-embedded system. The rate of review, discrepancies, doctor’s participation, and satisfaction were reported. The discrepancy scoring system derived from the RADPEER program.

Over a period of one year, 1,691 (1.1%) of 155,879 cases were reviewed: 89.7% (1,517) were qualified as concordant, 6.0% (102) as a minor discrepancy, 4.3% (72) as a major discrepancy, and 3.4% (54) as major discrepancies likely to be clinically significant.

52 of 65 radiologists (80%) participated in the review. 85% of them were overall satisfied with the peer review system, 88% felt that being reviewed was beneficial for them, 48% felt that reviewing others was beneficial for them.

All discrepancies likely to be clinically significant were consensually revised and the updated conclusions shared with the referring clinicians.

We highlight the feasibility of peer review on a voluntary basis within our network of teleradiologists. The recorded rate of discrepancies likely to be clinically significant was below the rates previously published (4). Participation and satisfaction among radiologists were high.

We therefore recommend the continuation of peer review in our community and encourage other teams to do so.

n/a

RIS-embedded system approuved by the CNIL (Comité national d'Informatique et des Libertées).

No funding was received for this work.

Purpose:

To present our experience and promote the learning and utilisation of a needs-based innovation approach in clinical radiology.

Need-based innovation is a methodology enabling a structured approach to screening for problems worth pursuing followed by a thorough system for idea generation and product development.

Oxford Biodesign offers two programmes which focus on a systematic approach to clinical-needs finding. Fellowship is a 9 month full immersion course taking participants from observations, through clinical needs finding, needs assessment, generation of solutions, stakeholders assessment, and market research, all the way to prototyping and start-up with seed funding. The short programme follows a similar curriculum, however, it is run as a series of two-hour interactive sessions once a week during the academic year.

Needs-based innovation is a systematic approach to problem-finding and solution generation which is well-tailored to the complex and high-barrier entry field of medical device and digital health industries. Clinician participation in the programme not only enables the team to benefit from the clinical insight but also enables the clinician to experience a broader view of healthcare and different cognitive approach to the clinical environment. In this presentation, we will demonstrate our project outcomes and their potential impact on clinical practice. In addition, we showcase our experience of participation in both of the programmes as well as the road to innovation in radiology.

Radiology is a rapidly developing, technology-based clinical speciality. Use of the systematic need led innovation is an effective way to guide radiologists and develop collaborations focused on medical innovation, including digital health and medical device development.

n/a

n/a

The Oxford Biodesign Programme is supported by institutions such as the European Institute of Innovation and Technology (EIT).

Purpose:

More than half of the world’s population lacks adequate medical imaging services according to the World Health Organisation (WHO). However, radiology plays an important role in public health initiative programs such as tuberculosis, trauma, breast cancer screening, and maternal-infant health. The purpose of this study was to identify challenges, opportunities, and strategies of radiology in global health initiatives in low-middle income country where resources may be scarce.

White paper reports of RAD-AID conferences on international radiology for developing countries since inception in 2009 were reviewed. Additionally, resources from national and international organisations that published data such as the World Bank and the World Health Organisation (WHO) were reviewed, as well as the medical literature databases from the Cochrane Library and PubMed. Search terms such as global health radiology, challenges, and strategies were used.

Establishment and sustainability of medical imaging services was a major long-standing challenge of radiology global health initiative programs in many low-income countries. To overcome this barrier, certain components of sustainability were identified as strategies to the implementation and optimisation of radiology services including economic development, technological innovation, clinical imaging models’ implementation, educational approaches, and the integration of public health to radiology.

Components of radiology global health initiative programs should consider economic sustainability, information technology innovation, tested clinical models, education, and public health policies as keys to optimising radiological services in low middle-income countries.

No known bias.

n/a

No funding was received for this work.

Purpose:

To evaluate user acceptance related to the RIS-PACS implementation in a clinical imaging department. To evaluate the degree to which the 3 constructs (deduced value, appreciated operational simplicity, and change) influence user acceptance and to assess whether socio-demographic variables explain user acceptance.

User acceptance levels among employees of the clinical imaging department at Al-Khor Hospital were assessed using a modified "model for adoption of technology (MAT)". 54 RIS-PACS questionnaire surveys were put in place to obtain user characteristics data, 6 items for deduced value (DV), 4 items for appreciated operational simplicity (AOS), 4 items for change construct, and 9 items for behaviour (acceptance) construct. Each item on the list used a five-point Likert scale for grading.

DV scored highly (4.31 out of 5), AOS 4.14 points, change scored 4.27, and acceptance level scored 3.88 points. Three constructs explained 42% of RIS-PACS variation in user acceptance, the most significant predictor being change construct explained by 32% of user variation. Users stated that the new RIS-PACS had a great impact on their work as it significantly improved overall service quality. Gender, age, profession, and frequency of RIS-PACS use did not alter the acceptance level. The influence of culture on the level of acceptance of RIS-PACS did not appear to have any influence.

Prior to RIS-PACS implementation, user acceptance should be assessed, as better user acceptance boosts chances of a successful adoption. Training programs are advised prior to its implementation. Healthcare organisations should measure all given factors influencing RIS-PACS acceptance. This will optimise system productivity and help in the realisation of its vast benefits.

Reasons for low acceptance levels among particular professional groups and other promising predictors of user acceptance need to be explored.

Ethics committee approval obtained.

No funding was received for this work.