[The contribution of artificial intelligence (AI) subsequent to the processing of thoracic imaging]. / Apport de l'intelligence artificielle dans le post-traitement de l'imagerie thoracique.

The contribution of artificial intelligence (AI) to medical imaging is currently the object of widespread experimentation. The development of deep learning (DL) methods, particularly convolution neural networks (CNNs), has led to performance gains often superior to those achieved by conventional methods such as machine learning. Radiomics is an approach aimed at extracting quantitative data not accessible to the human eye from images expressing a disease. The data subsequently feed machine learning models and produce diagnostic or prognostic probabilities. As for the multiple applications of AI methods in thoracic imaging, they are undergoing evaluation. Chest radiography is a practically ideal field for the development of DL algorithms able to automatically interpret X-rays. Current algorithms can detect up to 14 different abnormalities present either in isolation or in combination. Chest CT is another area offering numerous AI applications. Various algorithms have been specifically formed and validated for the detection and characterization of pulmonary nodules and pulmonary embolism, as well as segmentation and quantitative analysis of the extent of diffuse lung diseases (emphysema, infectious pneumonias, interstitial lung disease). In addition, the analysis of medical images can be associated with clinical, biological, and functional data (multi-omics analysis), the objective being to construct predictive approaches regarding disease prognosis and response to treatment.

THORACIC INVOLVEMENT IN CONNECTIVE TISSUE DISEASES: RADIOLOGICAL PATTERNS AND FOLLOW-UP.

Connective tissue diseases (CTDs) are a heterogeneous group of idiopathic inflammatory diseases involving various organs. A thoracic involvement is frequent, and chest-CT represents the imaging technique of reference in its assessment. Pulmonary abnormalities related to CTDs are various; although several disease-specific aspects have been described, the two most clinically relevant complications are represented by interstitial lung disease and pulmonary arterial hypertension. The early identification of a thoracic involvement, with the adoption of specific therapies, can significantly change patient's prognosis. The aim of this article is to review the most common typical and atypical CT features of thoracic involvement occurring in CT, especially focusing on interstitial lung disease.

Thoracic involvement in systemic primary vasculitis: radiological patterns and follow-up.

Systemic primary vasculitides are rare idiopathic diseases causing an inflammatory injury to the vessel walls. A pulmonary involvement is frequent, and chest-CT is the imaging technique of reference in its assessment. An extremely wide variety of parenchymal, vascular and airways abnormalities, has been described and diagnosis can be challenging: knowledge of clinical data and a close cooperation with the referring physician is often crucial. The aim of this work is to describe the most common typical and atypical CT features of pulmonary vasculitis and their possible changes over time and therapy, focusing on the differential diagnosis with other inflammatory/infectious or neoplastic diseases.

[Erdheim-Chester disease]. / Maladie d'Erdheim-Chester.

Erdheim-Chester disease is a rare and orphan disease. Despite having been overlooked previously, numerous new cases have been diagnosed more recently. The number of Erdheim-Chester disease cases reported has increased substantially: more than 300 new cases have been published in the past 10 years. This situation is mainly a result of the generally better awareness among pathologists, radiologists, and clinicians of various aspects of this rare disease. The field has been particularly active in the last few years, with evidence of the efficacy of interferon-α, the description of a systemic pro-inflammatory cytokine signature, and most recently, reports of the dramatic efficacy of BRAF inhibition in severe, BRAF(V600E) mutation-associated cases of Erdheim-Chester disease. Also, BRAF mutations have been found in more than half of the patients with Erdheim-Chester disease who were tested. Detailed elucidation of the pathogenesis of the disease is likely to lead to the development of better targeted and more effective therapies.

PHYSIOLOGIC AND QUANTITATIVE COMPUTED TOMOGRAPHY DIFFERENCES BETWEEN CENTRILOBULAR AND PANLOBULAR EMPHYSEMA IN COPD.

BACKGROUND: The purpose was to define the differences between centrilobular (CLE) and panlobular emphysema (PLE) phenotypes in cigarette smokers with COPD by a combined qualitative-quantitative computed tomography (CT) analysis. METHODS: Chest CT scans of 116 cigarette smokers were visually scored by 22 chest radiologists and 29 pulmonologists in a single setting for the predominant emphysema phenotype (e.g. CLE or PLE) and automatically quantified for emphysema{% low attenuation area (LAA) ≤ -950 HU - %LAAinsp-950, gas trapping extent and bronchial metrics{wall area % for segmental (%WAsegm) and subsegmental (%WAsubsegm) bronchi}. These quantitative CT indexes were compared and related to FEV1, FEV1/FVC, and smoking history as stratified for emphysema phenotype. RESULTS: Although more frequent than CLE in GOLD stages 3 and 4 (p = 0.01), PLE was also scored in 38.2% of combined GOLD stages 1 and 2. PLE was positively associated with %LAAinsp-950 (OR = 1.18, 95% CI: 1.12 to 1.27, ß coefficient = 0.17, p = <0.0001) and negatively associated with pack-years of smoking (OR = 0.97, 95% CI: 0.95 to 0.99, ß coefficient = -0.02, p = 0.03). Both %WAsegm and %WAsubsegm were more strongly associated with FEV1% (R2 = 0.6 for both measures, p< 0.001) in CLE as compared to PLE (R2= 0.15, p = 0.02; R2 = 0.26, p< 0.001). CONCLUSIONS: PLE likely represents a more advanced phase of emphysema, which may also occur in earlier COPD stages and show different interplay with airway disease as compared to CLE.

Integrated diagnostics: proceedings from the 9th biennial symposium of the International Society for Strategic Studies in Radiology.

The International Society for Strategic Studies in Radiology held its 9th biennial meeting in August 2011. The focus of the programme was integrated diagnostics and massive computing. Participants discussed the opportunities, challenges, and consequences for the discipline of radiology that will likely arise from the integration of diagnostic technologies. Diagnostic technologies are increasing in scope, including advanced imaging techniques, new molecular imaging agents, and sophisticated point-of-use devices. Advanced information technology (IT), which is increasingly influencing the practice of medicine, will aid clinical communication and the development of "population images" that represent the phenotype of particular diseases, which will aid the development of diagnostic algorithms. Integrated diagnostics offer increased operational efficiency and benefits to patients through quicker and more accurate diagnoses. As physicians with the most expertise in IT, radiologists are well placed to take the lead in introducing IT solutions and cloud computing to promote integrated diagnostics. To achieve this, radiologists must adapt to include quantitative data on biomarkers in their reports. Radiologists must also increase their role as participating physicians, collaborating with other medical specialties, not only to avoid being sidelined by other specialties but also to better prepare as leaders in the selection and sequence of diagnostic procedures. Key Points • New diagnostic technologies are yielding unprecedented amounts of diagnostic information.• Advanced IT/cloud computing will aid integration and analysis of diagnostic data.• Better diagnostic algorithms will lead to faster diagnosis and more rapid treatment.

Focal dependent pleural thickening at MDCT: pleural lesion or functional abnormality?

PURPOSE: To describe the characteristics of reversible focal pleural thickenings (PTs) mimicking real plaques, that firstly suggest asbestos exposure or pleural metastasis; to propose an imaging strategy and propose an explanation for their mechanism of formation. PATIENTS AND METHODS: Retrospective review of data from 19 patients with PTs fitting the description of pleural plaques at chest computed tomography (CT) and presenting modifications (clearance or appearance) of at least one PT at an additional chest examination in prone position. RESULTS: A total of 152 PTs were recorded on the first chest CT examinations with a range of two to 19 pleural opacities per patient. All PTs had a posterior distribution in the lower lobes. On the additional acquisitions, 144 PTs disappeared. Seventeen patients presented complete regression of PTs and two patients presented persistence of eight PTs. CONCLUSION: Additional low dose acquisition in prone position should be performed in all patients presenting with focal PT in a dependent and basal location. This may allow to exclude a pleural plaque in case of asbestos exposure but also a pleural metastasis in oncologic patients. These reversible dependent PTs could be related to physiological focal accumulation of lymphatic fluid in subpleural area.

[Bronchial diseases: CT imaging features]. / Sémiologie de l'atteinte bronchique en scanner.

Multidetector row computed tomography (MDCT) is the imaging modality of reference for the diagnosis of bronchiectasis. MDCT may also detect a focal stenosis, a tumor or multiple morphologic abnormalities of the bronchial tree. It may orient the endoscopist towards the abnormal bronchi, and in all cases assess the extent of the bronchial lesions. The CT findings of bronchial abnormalities include anomalies of bronchial division and origin, bronchial stenosis, bronchial wall thickening, lumen dilatation, and mucoid impaction. The main CT features of bronchiectasis are increased bronchoarterial ratio, lack of bronchial tapering, and visibility of peripheral airways. Other bronchial abnormalities include excessive bronchial collapse at expiration, outpouchings and diverticula, dehiscence, fistulas, and calcifications.

[Computer-aided detection of lung nodules on thin collimation MDCT: impact on radiologists' performance]. / Evaluation de l'impact d'un système CAD sur la performance des radiologues pour la détection des nodules pulmonaires sur des examens scanographiques multicoupes du thorax.

OBJECTIVES: Evaluate the improvement in detecting lung nodules when using multidetector CT (MDCT) computer-assisted diagnosis (CAD). MATERIAL AND METHODS: Three radiologists (R1, R2, R3) with different levels of experience independently interpreted 30 MDCT examinations of the thorax taken for screening purposes, first without and then with CAD. The diagnosis was established by two of the three radiologists interpreting the images together, assisted by the CAD. RESULTS: The consensus reading identified 133 nodules, 61 (46%) of which were 4 mm or larger. The sensitivity values in the detection of nodules before and after using the CAD were 54% and 80% (R1), 38% and 71% (R2), and 70% and 88% (R3), respectively. When considering only the nodules that were 4 mm or larger, the sensitivity values varied before and after using the CAD, from 62% to 95% (R1), from 41% to 84% (R2), and from 74% to 92% (R3). By combining two by two the three radiologists' results obtained without the CAD, the sensitivity values were 65%, 83%, and 77%, respectively, for all the nodules, and 70%, 85%, and 77% for the nodules that were 4 mm or larger. The CAD induced a total of 105 false-positive results, with a mean of 3.5 per examination. CONCLUSION: The lung nodules missed by the radiologist can be detected if the CAD is used as a second reader. The CAD can be at least as beneficial as the use of a second independent reader.

Quantification of bronchial dimensions at MDCT using dedicated software.

This study aimed to assess the feasibility of quantification of bronchial dimensions at MDCT using dedicated software (BronCare). We evaluated the reliability of the software to segment the airways and defined criteria ensuring accurate measurements. BronCare was applied on two successive examinations in 10 mild asthmatic patients. Acquisitions were performed at pneumotachographically controlled lung volume (65% TLC), with reconstructions focused on the right lung base. Five validation criteria were imposed: (1) bronchus type: segmental and subsegmental; (2) lumen area (LA)>4 mm2; (3) bronchus length (Lg) > 7 mm; (4) confidence index - giving the percentage of the bronchus not abutted by a vessel - (CI) >55% for validation of wall area (WA) and (5) a minimum of 10 contiguous cross-sectional images fulfilling the criteria. A complete segmentation procedure on both acquisitions made possible an evaluation of LA and WA in 174/223 (78%) and 171/174 (98%) of bronchi, respectively. The validation criteria were met for 56/69 (81%) and for 16/69 (23%) of segmental bronchi and for 73/102 (72%) and 58/102 (57%) of subsegmental bronchi, for LA and WA, respectively. In conclusion, BronCare is reliable to segment the airways in clinical practice. The proposed criteria seem appropriate to select bronchi candidates for measurement.

Evaluation of air trapping at CT: comparison of continuous-versus suspended-expiration CT techniques.

PURPOSE: To compare thin-section computed tomographic (CT) scans obtained during suspended end expiration with helical CT scans obtained during continuous expiration for the assessment of air trapping. MATERIALS AND METHODS: Forty-nine patients with an airway disease were examined with suspended-end-expiration CT after a 6-8-second expiratory maneuver, which was followed with continuous-expiration CT during a 10-second expiratory maneuver. The extent of expiratory air-trapping areas was calculated by two observers by using a semiquantitative grid score. The relative decrease in attenuation in the areas of air trapping was evaluated with a visual continuous-scale score. RESULTS: Air trapping was noted in 36 and 35 patients with continuous-expiration CT and with suspended-end-inspiration CT, respectively. The extents of and relative attenuation decreases in air-trapping areas in patients with air-trapping areas on at least one expiratory CT scan increased significantly in scans obtained with continuous-expiration CT compared with those obtained with suspended-end-expiration CT, respectively, with mean extent scores of 0.24 +/- 0.20 (SD) and 0.18 +/- 0.20 (paired t test, P: =.001) respectively, and with mean relative contrast decrease scores of 0.35 +/- 0.23 and 0.27 +/- 0.23 (paired t test, P: =.007), respectively. CONCLUSION: When suspended-end-expiration CT images are ambiguous, complementary continuous-expiration CT can be used to improve the conspicuity and apparent extent of air trapping.

Transjugular versus percutaneous renal biopsy for the diagnosis of parenchymal disease: comparison of sampling effectiveness and complications.

PURPOSE: To compare the effectiveness and safety of transjugular renal biopsy with those of percutaneous renal biopsy for diagnosis of renal parenchymal disease. MATERIALS AND METHODS: Results and complications of 400 consecutive transjugular renal biopsies performed between 1993 and 1998 with a modified Colapinto transjugular hepatic biopsy system were compared retrospectively with those of 400 percutaneous renal biopsies performed during the same period. Transjugular renal biopsy was associated with 14 cardiac and 35 hepatic biopsies. Number of glomeruli per tissue core, adequacy of tissue core for histopathologic diagnosis, and rate and severity of complications were analyzed. RESULTS: Renal tissue was obtained with percutaneous renal biopsy in 382 (95.5%) of 400 patients and with transjugular renal biopsy in 383 (95.8%) of 400 patients. The mean numbers of intact glomeruli per tissue core with optical microscopy were 11.2 +/- 7.7 (SD) and 9.8 +/- 7.6 for percutaneous renal biopsy and transjugular renal biopsy, respectively. With immunofluorescent microscopy, the mean numbers were 6.4 +/- 5.3 and 4.6 +/- 4.6 for percutaneous renal biopsy and transjugular renal biopsy, respectively. Tissue cores were adequate for histopathologic diagnosis in 98.2% with both techniques. Major complications occurred with transjugular renal biopsy in four patients and with percutaneous renal biopsy in three patients. CONCLUSION: Use of transjugular renal biopsy provides diagnostic yield and safety similar to those of percutaneous renal biopsy and allows multiorgan biopsy during the same procedure. It can be recommended in patients with percutaneous renal biopsy contraindication or failure.

Diaphragm and chest wall: assessment of the inspiratory pump with MR imaging-preliminary observations.

Magnetic resonance (MR) imaging of the thorax with three-dimensional (3D) reconstruction and functional quantification was evaluated as a tool for structure-function evaluation of chest-wall mechanics. Good agreement was found between the corresponding spirometric and MR imaging values of lung volumes. Fast MR imaging of the thorax with 3D reconstruction should improve the ability to evaluate the inspiratory pump in clinical and research investigations.

MR arthrography of the hip: normal intra-articular structures and common disorders.

This pictorial review illustrates the anatomical features of normal intra-articular components of the hip and their common disorders on MR arthrography. On T1-weighted MR arthrograms, the normal contrast-filled joint cavity shows a homogeneous high signal intensity. Normal acetabular labrum appears as a well-delineated triangle showing a low signal intensity, surrounded by contrast material in the perilabral recess. Intra-articular paramagnetic contrast outlines labral tears, loose bodies, communicating labral cysts and cartilage lesions (traumatic tears, focal defects, degenerative fissures and thinning), and improves their detection. Overall, MR arthrography enables accurate detection and staging of hip intra-articular structure abnormalities.

Intrapulmonary lesions: percutaneous automated biopsy with a detachable, 18-gauge, coaxial cutting needle.

PURPOSE: To evaluate a lung biopsy technique in which a detachable, 18-gauge, coaxial guide around a central notched stylet is used as a cutting needle. MATERIALS AND METHODS: The records of 89 consecutive patients (41 women, 48 men; aged 21-86 years) who underwent coaxial percutaneous core biopsy of 91 lung lesions that required needle passage through normal lung tissue (mean lesion size, 33.6 mm; range, 9-80 mm) were studied. Detachable, 18-gauge, coaxial automated cutting needles were used. RESULTS: The mean number of needle passes was 2.5 (range, 1-4). All biopsies yielded sufficient tissue for histopathologic (n = 91) and, if necessary, bacteriologic (n = 12) evaluation (mean core length, 5 mm; range, 1-15 mm). Eighty-nine lesions had definitive diagnoses. Seventy-five lesions were proved to be malignant; seventy (93%) could be accurately diagnosed with coaxial percutaneous core biopsy samples. Fourteen lesions were proved to be benign; 10 (71%) were specifically diagnosed with biopsy samples. Among the 91 biopsies, the overall diagnostic accuracy was 88% (80 of 91 lesions). A pneumothorax occurred in 31 cases (34%), three (3%) of which necessitated placement of a chest tube. Postbiopsy hemoptysis occurred and resolved spontaneously in nine cases (10%). CONCLUSION: This technique provides a core biopsy specimen without the need for an on-site cytopathologist during the procedure. It has a high diagnostic accuracy and an acceptable rate of complications.

Spiral computed tomographic scanning and magnetic resonance angiography for the diagnosis of pulmonary embolism.

PURPOSE: To compare prospectively the accuracy of spiral computed tomography (CT) with that of ventilation-perfusion scintigraphy for diagnosing pulmonary embolism. MATERIALS AND METHODS: Within 48 hours of presentation, 142 patients suspected of having pulmonary embolism underwent spiral CT, scintigraphy, and (when indicated) pulmonary angiography. Pulmonary angiography was attempted if interpretations of spiral CT scans and of scintigrams were discordant or indeterminate and intermediate-probability, respectively. RESULTS: In the 139 patients who completed the study, interpretations of spiral CT scans and of scintigrams were concordant in 103 patients (29 with embolism, 74 without). In 20 patients, intermediate-probability scintigrams were interpreted (six with embolism at angiography, 14 without); diagnosis with spiral CT was correct in 16. Interpretations of spiral CT scans and those of scintigrams were discordant in 12 cases; diagnosis with spiral CT was correct in 11 cases and that with scintigraphy was correct in one. Spiral CT and scintigraphic scans of four patients with embolism did not show embolism. Sensitivities, specificities, and kappa values with spiral CT and scintigraphy were 87%, 95%, and 0.85 and 65%, 94%, and 0.61, respectively. CONCLUSION: In cases of pulmonary embolism, sensitivity of spiral CT is greater than that of scintigraphy. Interobserver agreement is better with spiral CT.