Artificial intelligence to support person-centred care in breast imaging - A scoping review.

AIM: To overview Artificial Intelligence (AI) developments and applications in breast imaging (BI) focused on providing person-centred care in diagnosis and treatment for breast pathologies. METHODS: The scoping review was conducted in accordance with the Joanna Briggs Institute methodology. The search was conducted on MEDLINE, Embase, CINAHL, Web of science, IEEE explore and arxiv during July 2022 and included only studies published after 2016, in French and English. Combination of keywords and Medical Subject Headings terms (MeSH) related to breast imaging and AI were used. No keywords or MeSH terms related to patients, or the person-centred care (PCC) concept were included. Three independent reviewers screened all abstracts and titles, and all eligible full-text publications during a second stage. RESULTS: 3417 results were identified by the search and 106 studies were included for meeting all criteria. Six themes relating to the AI-enabled PCC in BI were identified: individualised risk prediction/growth and prediction/false negative reduction (44.3%), treatment assessment (32.1%), tumour type prediction (11.3%), unnecessary biopsies reduction (5.7%), patients' preferences (2.8%) and other issues (3.8%). The main BI modalities explored in the included studies were magnetic resonance imaging (MRI) (31.1%), mammography (27.4%) and ultrasound (23.6%). The studies were predominantly retrospective, and some variations (age range, data source, race, medical imaging) were present in the datasets used. CONCLUSIONS: The AI tools for person-centred care are mainly designed for risk and cancer prediction and disease management to identify the most suitable treatment. However, further studies are needed for image acquisition optimisation for different patient groups, improvement and customisation of patient experience and for communicating to patients the options and pathways of disease management.

A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging.

PURPOSE: To review eXplainable Artificial Intelligence/(XAI) methods available for medical imaging/(MI). METHOD: A scoping review was conducted following the Joanna Briggs Institute's methodology. The search was performed on Pubmed, Embase, Cinhal, Web of Science, BioRxiv, MedRxiv, and Google Scholar. Studies published in French and English after 2017 were included. Keyword combinations and descriptors related to explainability, and MI modalities were employed. Two independent reviewers screened abstracts, titles and full text, resolving differences through discussion. RESULTS: 228 studies met the criteria. XAI publications are increasing, targeting MRI (n = 73), radiography (n = 47), CT (n = 46). Lung (n = 82) and brain (n = 74) pathologies, Covid-19 (n = 48), Alzheimer's disease (n = 25), brain tumors (n = 15) are the main pathologies explained. Explanations are presented visually (n = 186), numerically (n = 67), rule-based (n = 11), textually (n = 11), and example-based (n = 6). Commonly explained tasks include classification (n = 89), prediction (n = 47), diagnosis (n = 39), detection (n = 29), segmentation (n = 13), and image quality improvement (n = 6). The most frequently provided explanations were local (78.1 %), 5.7 % were global, and 16.2 % combined both local and global approaches. Post-hoc approaches were predominantly employed. The used terminology varied, sometimes indistinctively using explainable (n = 207), interpretable (n = 187), understandable (n = 112), transparent (n = 61), reliable (n = 31), and intelligible (n = 3). CONCLUSION: The number of XAI publications in medical imaging is increasing, primarily focusing on applying XAI techniques to MRI, CT, and radiography for classifying and predicting lung and brain pathologies. Visual and numerical output formats are predominantly used. Terminology standardisation remains a challenge, as terms like "explainable" and "interpretable" are sometimes being used indistinctively. Future XAI development should consider user needs and perspectives.

Plain radiography has a role to play in current clinical practice in Western Switzerland.

AIM: The aim of the study was to investigate the current role of conventional radiography examinations in Western Switzerland and the main clinical indications required to justify the use of this imaging examination. METHODS: Ethical approval was obtained from Vaud Ethics committee (Ref 2020-00311). An online questionnaire was specifically designed and implemented on the data collection tool LimeSurvey composed of two parts: a) to characterise the participants' profile and their institutions and b) 169 projections for the different anatomical area (upper and lower limbs, pelvis, skull, spine, thorax, abdomen) were presented to collect data about the frequency and main clinical indications. Statistical analysis was performed using the software IBM SPSS® (Statistical Package for the Social Sciences) version 26. RESULTS: Radiographers from 60% (26/43) of the invited institutions participated in this survey, mainly from Vaud region. The upper and lower limbs were the most commonly examined by using conventional radiography mainly for trauma and degenerative disorders. The thorax was also an anatomical area commonly explored by X-rays, so were the spine (cervical and lumbar lateral). The skull radiographs were rarely performed in clinical practice and some of the projections were not being used, namely Hirtz, Tangential Nose Bones, Worms and Caldwell's views. CONCLUSIONS: Plain radiography is being used in clinical practice mainly for appendicular skeleton studies and for trauma and degenerative pathologies. Adaptations in radiographers' education and training and other healthcare professionals are needed to provide the judicious use of data that radiographs can give to better manage the patients' imaging pathway.

Feasibility and Performance of Free-Hand Single-Photon Computed Tomography/Ultrasonography for Preoperative Parathyroid Adenoma Localization: A Pilot Study.

The aim of this prospective pilot study was to evaluate the feasibility of a new hybrid imaging modality, free-hand single-photon computed tomography/ultrasonography (fhSPECT/US), for preoperative localization of parathyroid adenomas and to compare its performance with conventional ultrasonography and SPECT/CT. Twelve patients diagnosed with primary hyperparathyroidism underwent sequentially US and parathyroid scintigraphy, including SPECT/CT, followed by fhSPECT/US, allowing for real-time fusion between US and freehand-generated gamma-camera images. The fhSPECT/US detection rates were correlated with histopathology, when available, or with the imaging modality showing the most lesions. Based on a per patient analysis, the detection rate was significantly different when comparing SPECT/CT to fhSPECT/US (p = 0.03), and not significantly different when comparing SPECT/CT to US (p = 0.16) and US to fhSPECT/US (p = 0.08). Based on a per-lesion analysis, the detection rate of SPECT/CT was significantly higher than that of US (p = 0.01) and fhSEPCT/US (p = 0.003), and there was no significant difference in detection rate when comparing US to fhSPECT/US (p = 0.08). The main perceived limitations of fhSPECT/US in lesion detection were: (i) lesions localized at a depth ≥4.5 cm; (ii) imperfect image fusion due to tissue compression; (iii) limited spatial manipulation ability of the SPECT mobile camera handheld probe; and (iv) a wide spread of detected activity. In conclusion, clinical use of fhSPECT/US for localization of parathyroid adenomas is feasible, but shows lower sensitivity than conventional modalities and requires technical improvements.

Biokinetics and dosimetry of 111In-DOTA-NOC-ATE compared with 111In-DTPA-octreotide.

PURPOSE: The biokinetics and dosimetry of (111)In-DOTA-NOC-ATE (NOCATE), a high-affinity ligand of SSTR-2 and SSTR-5, and (111)In-DTPA-octreotide (Octreoscan™, OCTREO) were compared in the same patients. METHODS: Seventeen patients (10 men, 7 women; mean age 60 years), referred for an OCTREO scan for imaging of a neuroendocrine tumour (15), thymoma (1) or medullary thyroid carcinoma (1), agreed to undergo a second study with NOCATE. Whole-body anterior-posterior scans were recorded 0.5 (100 % reference scan), 4, 24 and 48 h (17 patients) and 120 h (5 patients) after injection. In 16 patients the OCTREO scan (178 ± 15 MBq) was performed 16 ± 5 days before the NOCATE scan (108 ± 14 MBq) with identical timing; 1 patient had the NOCATE scan before the OCTREO scan. Blood samples were obtained from 14 patients 5 min to 48 h after injection. Activities expressed as percent of the initial (reference) activity in the whole body, lung, kidney, liver, spleen and blood were fitted to biexponential or single exponential functions. Dosimetry was performed using OLINDA/EXM. RESULTS: Initial whole-body, lung and kidney activities were similar, but retention of NOCATE was higher than that of OCTREO. Liver and spleen uptakes of NOCATE were higher from the start (p < 0.001) and remained so over time. Whole-body activity showed similar α and ß half-lives, but the ß fraction of NOCATE was double that of OCTREO. Blood T (1/2)ß for NOCATE was longer (19 vs. 6 h). As a result, the effective dose of NOCATE (105 µSv/MBq) exceeded that of OCTREO (52 µSv/MBq), and the latter result was similar to the ICRP 106 value of 54 µSv/MBq. Differential activity measurement in blood cells and plasma showed an average of <5 % of NOCATE and OCTREO attached to globular blood components. CONCLUSION: NOCATE showed a slower clearance from normal tissues and its effective dose was roughly double that of OCTREO.

EXPOSURE OF THE SWISS POPULATION BY RADIODIAGNOSTICS: 2013 REVIEW.

In 2013, a nationwide investigation was conducted in Switzerland to establish the population's exposure from medical X rays. A hybrid approach was used combining the Raddose database accessible on-line by the participating practices and the Swiss medical tariffication system for hospitals. This study revealed that the average annual number of examinations is 1.2 per inhabitant, and the associated annual effective dose is 1.4 mSv. It also showed that computed tomography is the most irradiating modality and that it delivers 70 % of the total dose. The annual effective dose per inhabitant registered a 17 % increase in 5 y and is comparable with what was recently reported in neighbouring countries.