Autonomous artificial intelligence in pediatric radiology: the use and perception of BoneXpert for bone age assessment.

BACKGROUND: The autonomous artificial intelligence (AI) system for bone age rating (BoneXpert) was designed to be used in clinical radiology practice as an AI-replace tool, replacing the radiologist completely. OBJECTIVE: The aim of this study was to investigate how the tool is used in clinical practice. Are radiologists more inclined to use BoneXpert to assist rather than replace themselves, and how much time is saved? MATERIALS AND METHODS: We sent a survey consisting of eight multiple-choice questions to 282 radiologists in departments in Europe already using the software. RESULTS: The 97 (34%) respondents came from 18 countries. Their answers revealed that before installing the automated method, 83 (86%) of the respondents took more than 2 min per bone age rating; this fell to 20 (21%) respondents after installation. Only 17/97 (18%) respondents used BoneXpert to completely replace the radiologist; the rest used it to assist radiologists to varying degrees. For instance, 39/97 (40%) never overruled the automated reading, while 9/97 (9%) overruled more than 5% of the automated ratings. The majority 58/97 (60%) of respondents checked the radiographs themselves to exclude features of underlying disease. CONCLUSION: BoneXpert significantly reduces reporting times for bone age determination. However, radiographic analysis involves more than just determining bone age. It also involves identification of abnormalities, and for this reason, radiologists cannot be completely replaced. AI systems originally developed to replace the radiologist might be more suitable as AI assist tools, particularly if they have not been validated to work autonomously, including the ability to omit ratings when the image is outside the range of validity.

Identifying existential concerns as an analytical tool: an eye-opener for the doctor.

Existential concerns manifest themselves emotionally in patients. Emotions tend to be transferred between patient and doctor but the underlying existential concerns may remain hidden and obscure for both. If doctors understand that there are always existential concerns behind patients' inquiries it becomes easier to relate to the patient's feelings in an interested and curious way. Ultimately this benefits both doctors and patients. We have observed five existential human concerns leads to strong emotions (1. death, 2. thrown-ness, 3. aloneness, 4. choosing (the imperative of choice) and 5. meaninglessness (the absence of objective meaning).

Helping doctors to improve the 'Patient's Part' of consultation using the 'Macro-Micro Supervision' teaching method.

A doctor-patient consultation can be viewed as consisting of three parts: Patient's Part, Doctor's Part and Shared Part. Macro-Micro Supervision is a teaching method developed to train doctors in consultation techniques for the initial Patient's Part and to become more patient-centred. Doctors find the Patient's Part of our consultation method the most difficult. Macro-Micro Supervision is used when groups review video-recorded role-plays of real consultations or of simulated consultations, where participants alternate between playing doctor and patient. It can also be used in one-to-one supervision.

Preventing contrast medium-induced acute kidney injury : Side-by-side comparison of Swedish-ESUR guidelines.

A side-by-side comparison of updated guidelines regarding contrast medium-induced acute kidney injury (CI-AKI) from the Swedish Society of Uroradiology (SSUR) and the European Society of Urogenital Radiology (ESUR) is presented. The major discrepancies include a higher glomerular filtration rate (GFR) threshold as a risk factor for CI-AKI and for discontinuation of metformin by SSUR, i.e., < 45 ml/min versus < 30 ml/min/1.73 m2 by ESUR, when intravenous or intra-arterial contrast media (CM) with second-pass renal exposure is administered. SSUR also continues to recommend consideration of traditional non-renal risk factors such as diabetes and congestive heart failure, while ESUR considers these factors as non-specific for CI-AKI and does not recommend any consideration. Contrary to ESUR, SSUR also recommends discontinuation of NSAID and nephrotoxic medication if possible. Insufficient evidence at the present time motivates the more cautionary attitude taken by SSUR. Furthermore, SSUR expresses GFR thresholds in absolute values in ml/min as recommended by the National Kidney Foundation for drugs excreted by glomerular filtration, while ESUR uses the relative GFR normalised to body surface area in ml/min/1.73 m2. CM dose/GFR ratio thresholds established for coronary angiography/interventions are also applied as recommendations for CM-enhanced CT by SSUR, since SSUR regards coronary procedures as a second-pass renal exposure of CM with no obvious difference in the incidence of AKI compared with IV CM administration. Finally, SSUR recommends reducing the gram-iodine dose/GFR ratio from < 1.0 in patients not at risk to < 0.5 in patients at risk of CI-AKI, while ESUR has no such recommendation. KEY POINTS: • The more cautionary attitude taken by SSUR compared with that of ESUR is motivated by insufficient evidence regarding risk for contrast medium-induced acute kidney injuries (CI-AKI). • SSUR recommends that absolute and not relative GFR should be used when dosing drugs eliminated by the kidneys such as contrast media. • According to SSUR the gram-iodine dose/GFR ratio should be < 0.5 in patients at risk of CI-AKI, while ESUR has no such recommendation.