Machine vs. Radiologist-Based Translations of RadLex: Implications for Multi-language Report Interoperability

The purpose of this study was to evaluate the feasibility of translation of RadLex lexicon from English to German performed by Google Translate, using the RadLex ontology as ground truth. The same comparison was also performed for German to English translations. We determined the concordance rate of the Google Translate-rendered translations (for both English to German and German to English translations) to the official German RadLex (translations provided by the German Radiological Society) and English RadLex terms via character-by-character concordance analysis (string matching). Specific term characteristics of term character count and word count were compared between concordant and discordant translations using t-tests. Google Translate-rendered translations originally considered incongruent (2482 English terms and 2500 German terms) were then reviewed by German and English-speaking radiologists to further evaluate clinical utility. Overall success rates of both methods were calculated by adding the percentage of terms marked correct by string comparison to the percentage marked correct during manual review extrapolated to the terms that had been initially marked incorrect during string analysis. 64,632 English and 47,425 German RadLex terms were analyzed. 3507 (5.4%) of the Google Translate-rendered English to German translations were concordant with the official German RadLex terms when evaluated via character-by-character concordance. 3288 (6.9%) of the Google Translate-rendered German to English translations matched the corresponding English RadLex terms. Human review of a random sample of non-concordant machine translations revealed that 95.5% of such English to German translations were understandable, whereas 43.9% of such German to English translations were understandable. Combining both string matching and human review resulted in an overall Google Translate success rate of 95.7% for English to German translations and 47.8% for German to English translations. For certain radiologic text translation tasks, Google Translate may be a useful tool for translating multi-language radiology reports into a common language for natural language processing and subsequent labeling of datasets for machine learning. Indeed, string matching analysis alone is an incomplete method for evaluating machine translation. However, when human review of automated translation is also incorporated, measured performance improves. Additional evaluation using longer text samples and full imaging reports is needed. An apparent discordance between English to German versus German to English translation suggests that the direction of translation affects accuracy.

Structured reporting in radiology: a systematic review to explore its potential

OBJECTIVES

Structured reporting (SR) in radiology reporting is suggested to be a promising tool in clinical practice. In order to implement such an emerging innovation, it is necessary to verify that radiology reporting can benefit from SR. Therefore, the purpose of this systematic review is to explore the level of evidence of structured reporting in radiology. Additionally, this review provides an overview on the current status of SR in radiology.

METHODS

A narrative systematic review was conducted, searching PubMed, Embase, and the Cochrane Library using the syntax 'radiol*' AND 'structur*' AND 'report*'. Structured reporting was divided in SR level 1, structured layout (use of templates and checklists), and SR level 2, structured content (a drop-down menu, point-and-click or clickable decision trees). Two reviewers screened the search results and included all quantitative experimental studies that discussed SR in radiology. A thematic analysis was performed to appraise the evidence level.

RESULTS

The search resulted in 63 relevant full text articles out of a total of 8561 articles. Thematic analysis resulted in 44 SR level 1 and 19 level 2 reports. Only one paper was scored as highest level of evidence, which concerned a double cohort study with randomized trial design.

CONCLUSION

The level of evidence for implementing SR in radiology is still low and outcomes should be interpreted with caution.

KEY POINTS

• Structured reporting is increasingly being used in radiology, especially in abdominal and neuroradiological CT and MRI reports. • SR can be subdivided into structured layout (SR level 1) and structured content (SR level 2), in which the first is defined as being a template in which the reporter has to report; the latter is an IT-based manner in which the content of the radiology report can be inserted and displayed into the report. • Despite the extensive amount of research on the subject of structured reporting, the level of evidence is low.

Analysis of news media reports of suicides and attempted suicides during the COVID-19 lockdown in India

BACKGROUND

Based on previous experience there is justifiable concern about suicidal behaviour and news media reporting of it during COVID-19 pandemic.

METHODS

This study used a systematic search of online news media reports (versions of newspapers, magazine and other digital publications) of suicidal behaviour during India's COVID-19 lockdown and compared it to corresponding dates in 2019. Data was gathered using a uniform search strategy from 56 online news media publications 24 March to 3 May for the years 2019 and 2020 using keywords, suicide, attempted suicide, hangs self and kills self. Demographic variables and methods used for suicide were compared for suicide and attempts between the 2 years using chi-squared tests (χ2).

RESULTS

There were online news media reports of 369 cases of suicides and attempted suicides during COVID lockdown vs 220 reports in 2019, a 67.7% increase in online news media reports of suicidal behaviour. Compared to 2019, suicides reported during lockdown were significantly older (30 vs 50 years, p < 0.05), men (71.2% vs 58.7%; p < 0.01), married (77.7% vs 49%; p < 0.01) and employed (82.9% vs 59.5%; p < 0.01). During the lockdown, significantly more suicides were by hanging (64.4% vs 42%), while poisoning (8.5% vs 21.5%) and jumping in front of a train (2% vs 9.4%) (p < 0.05) were significantly reduced. Comparison of COVID and non-COVID groups showed that online news media reports of COVID cases of suicide and attempted suicide were significantly more likely to be men (84.7% vs 60.4%; p < 0.01), older (31-50 years 52.9% vs 25.8%; p < 0.01) employed (91.5% vs 64.3%; p < 0.01), had poor mental (40.1% vs 20.8%; p < 0.01) and poor physical health (24.8% vs 7.9%;11.8, p < 0.01).

CONCLUSION

Increase in online news media reports of suicides and attempts during COVID-19 lockdown may indicate an increase in journalists' awareness about suicide or more sensational media reporting or may be a proxy indicator of a real community increase in suicidal behaviour. It is difficult to attribute changes in demographic profile and methods used only to changes in journalists' reporting behaviour and should be further explored. We therefore call upon the Government of India to urgently release national suicide data to help devise a comprehensive suicide prevention strategy to address COVID-19 suicidal behaviour.

Ensuring trial conduct is consistent with trial design: assumption is the enemy of quality

‘Assumptions are made and most assumptions are wrong’ (Albert Einstein)

Clinical trial conduct must be consistent with trial design, yet conducting the trial according to plan remains a major challenge.

We discuss the importance of optimal co-applicant team formation in trial leadership, appropriate delegation of tasks and staff supervision arrangements. Finally, we discuss five standard documents which we believe require particular attention. With appropriate engagement by or with co-applicants during the preparation of these five standard documents, we believe many of the pitfalls trials commonly experience can be avoided. The risks inherent in failing to identify and address mistaken assumptions during the preparation of these documents are discussed and recommendations for best practice suggested.

Toward an Understanding of the Essential Components of Behavior Analytic Service Plans

Baer, Wolf, and Risley (1968) indicated "technological" was one of seven core dimensions of applied behavior analysis (ABA). They described this dimension as being met if interventions were described well enough to be implemented correctly. Often in the applied settings, a behavior plan is the method by which interventions are communicated to staff and parents for implementation. The necessary components of a behavior plan have been discussed in relation to compliance with regulations (e.g., Vollmer, Iwata, Zarcone, & Rodgers, Research in Developmental Disabilities 13:429-441, 1992), in school settings (e.g., Horner, Sugai, Todd, & Lewis-Palmer, Exceptionality: A Special Education Journal 8:205-215, 2000), and other applied settings (e.g., Tarbox et al., Research in Autism Spectrum Disorder 7:1509-1517, 2013) for the last 25 years. The purpose of this research is to review the literature regarding components of behavior plans and synthesize it with a recent survey of behavior analysts regarding essential components of behavior plans. The results are discussed in light of training, treatment fidelity implications (i.e., Registered Behavior Technician Task List), public policy development (e.g., state initiative for a single behavior plan template), and research opportunities (e.g., comparison of different visual structures).

Use of a trigger tool to detect adverse drug reactions in an emergency department

Background

Although there are systems for reporting adverse drug reactions (ADR), these safety events remain under reported. The low-cost, low-tech trigger tool method is based on the detection of events through clues, and it seems to increase the detection of adverse events compared to traditional methodologies. This study seeks to estimate the prevalence of adverse reactions to drugs in patients seeking care in the emergency department.

Methods

Retrospective study from January to December, 2014, applying the Institute for Healthcare Improvement (IHI) trigger tool methodology for patients treated at the emergency room of a tertiary care hospital.

Results

The estimated prevalence of adverse reactions in patients presenting to the emergency department was 2.3% [CI₉₅ 1.3% to 3.3%]; 28.6% of cases required hospitalization at an average cost of US$ 5698.44. The most common triggers were hydrocortisone (57% of the cases), diphenhydramine (14%) and fexofenadine (14%). Anti-infectives (19%), cardiovascular agents (14%), and musculoskeletal drugs (14%) were the most common causes of adverse reactions. According to the Naranjo Scale, 71% were classified as possible and 29% as probable. There was no association between adverse reactions and age and sex in the present study.

Conclusions

The use of the trigger tool to identify adverse reactions in the emergency department was possible to identify a prevalence of 2.3%. It showed to be a viable method that can provide a better understanding of adverse drug reactions in this patient population.

Quality indicators for esophagogastroduodenoscopy: A comparative study of outcomes after an improvement programme in a tertiary hospital

INTRODUCTION

There is an opportunity for improvement in the recording and measuring of quality indicators. However, no previous experiences exist in our field in terms of their compliance in esophagogastroduodenoscopy (EGD).

OBJECTIVE

To analyse compliance with EGD quality criteria and evaluate improvement after conducting a training programme.

PATIENTS AND METHODS

Comparative study of 2 cohorts: one retrospective (control group) and one prospective (intervention group), before and after a training programme consisting of an information session and the report writing improvement programme. The quality indicators proposed by the American Society for Gastrointestinal Endoscopy and the American College of Gastroenterology were used.

RESULTS

A total of 1,200 EGDs were included in a sequential manner (600 in each group). Following the training programme, a significant improvement was observed in the following indicators: documented indication (93 vs. 99.8%; P<0.01), documented full examinations (94.7 vs. 97.3%; P<0.01), correct performance (63.7 vs. 87.9%; P<0.01), appropriate biopsies according to protocols (57.9 vs. 83.8%; P<0.01), photo-documentation of described lesions (84.1 vs. 94.9%; P<0.01), photo-documentation per segment (52.9 vs. 70.5%; P<0.01) and correct overall assessment (56,9 vs. 90.5%; P<0.01). Biopsies for coeliac disease, documented indication, full examination and correct performance, if it went ahead, exceeded the recommended standard.

CONCLUSION

A very simple training programme improves EGD quality indicators, with the majority reaching the standards recommended by the American Society for Gastrointestinal Endoscopy and the American College of Gastroenterology.

[Clinically relevant incidental cardiovascular findings in CT examinations]

BACKGROUND AND METHOD

Incidental cardiovascular findings are a frequent phenomenon in computed tomography (CT) examinations. As the result of a dedicated PubMed search this article gives a systemic overview of the current literature on the most important incidental cardiovascular findings, their prevalence and clinical relevance.

RESULTS

The majority of incidental cardiovascular findings are of only low clinical relevance; however, highly relevant incidental findings, such as aortic aneurysms, thromboses and thromboembolic events can also occasionally be found, especially in oncology patients.

CONCLUSION

The scans from every CT examination should also be investigated for incidental findings as they can be of decisive importance for the further clinical management of patients, depending on their clinical relevance.

Quality assessment of pelvic ultrasound for uterine myoma according to the CNGOF guidelines

INTRODUCTION

French guidelines regarding the minimum criteria for gynaecological ultrasound were given in a recent report in 2016, by the French National College of Obstetricians and Gynaecologists (CNGOF). An accurate report is essential for the optimal care of women, especially those presenting myomas. The goal of this study was to evaluate the quality of gynaecological ultrasound reports for women with type 0 to 2 uterine myomas, referring to the items contained in the French guidelines.

MATERIALS AND METHODS

A retrospective descriptive study was conducted from reports of ultrasounds performed in private offices and in the gynaecologic department of a hospital, between June 2014 and June 2016 (before the report of CNGOF). These reports involved women who underwent hysteroscopic resection of myoma(s). A search of validated items was conducted for all of the reports, and the missing items were analysed. The different types of practitioners and between hospital and private medical offices were also compared with Chi-square tests.

RESULTS

A total of 138 reports were analysed; 71 were performed in private offices and 67 were performed in the gynaecologic unit of the hospital. Many items were missing in the reports, with disparities between the type of institution (private offices or hospital) and the speciality of practitioners (radiologists or gynaecologists). Specific items regarding myomas, such as the International Federation of Gynaecologists and Obstetricians (FIGO) classification or measurement of the posterior wall, were more often missing in reports from radiologists (89.7% and 79.5%, respectively) than in reports from gynaecologists (21.2% and 34.3%, respectively) (P<0.05). A significant difference was also observed for these data between private offices' reports and hospitals' reports. Items relative to ultrasound structures, such as the appearance of myomas or associated abdominal effusion, were more frequently missing in gynaecologists' reports (88.9% and 49.5%, respectively) compared to radiologists' reports (56.4% and 12.8%, respectively) (P<0.05).

CONCLUSIONS

Certain items are present in all the reports, while others are insufficiently mentioned. These inequalities can be explained in part by the type of practice; however, methods to overcome these difficulties must be developed. Information campaigns to educate professionals on the minimum reporting and training conducted jointly by radiologists and gynaecologist surgeons might improve reports and improve the care of women.

Modeling a System for Generating Structured Reports

The purpose of this research is to make the medical report generation process more practical, fast and reliable, both for the health professional and for the patient. We created an ontology and modeling of a structured report (SR) Standard DICOM SR.

Are amended surgical pathology reports getting to the correct responsible care provider?

OBJECTIVES

Amended reports (AmRs) need to follow patients to treating physicians, to avoid erroneous management based on the original diagnosis. This study was undertaken to determine if AmRs followed the patient appropriately.

METHODS

AmRs with diagnostic changes and discrepancies between ordering and treating physicians were tracked. Chart reviews, electronic medical report (EMR) reviews, and interviews were conducted to establish receipt of the AmR by the correct physician.

RESULTS

Seven of 60 AmRs had discrepancies between the ordering and treating physicians, all with malignant diagnoses. The AmR was present in the treating physician's chart in only one case. Ordering physicians indicated that AmRs were not forwarded to treating physicians when corrected results arrived after patient referral, under the assumption that the new physician was automatically forwarded pathology updates. No harm was documented in any of our cases. In one case with a significant amendment, the correct information was entered in the patient chart based on a tumor board discussion. A review of two electronic health record systems uncovered significant shortcomings in each delivery system.

CONCLUSIONS

AmRs fail to follow the patient's chain of referrals to the correct care provider, and EMR systems lack the functionality to address this failure and alert clinical teams of amendments.

Cautionary note: Electronic medical records, a potential disaster in the making?

Concern is expressed that electronic medical records may actually compromise care. Reports are electronically collated with patient charts, but when are they examined? Current electronic transmission of results to patients’ electronic medical records do not seem to notify of new information. The unknown time from prescription to patient action and the variable time required for individual test performance seem to mandate that a physician attempting to be conscientious would have to examine all sections of every patient medical record in their practice, every day. That is quite inefficient and error-prone. Electronic medical record still contains what appear to be dangerous “bugs” which compromise our ability to provide the care we believe our patients deserve? I remain unsure that outpatient electronic medical records are “ready for prime time.”