Assessment of Radiology Artificial Intelligence Software: A Validation and Evaluation Framework.

Artificial intelligence (AI) software in radiology is becoming increasingly prevalent and performance is improving rapidly with new applications for given use cases being developed continuously, oftentimes with development and validation occurring in parallel. Several guidelines have provided reporting standards for publications of AI-based research in medicine and radiology. Yet, there is an unmet need for recommendations on the assessment of AI software before adoption and after commercialization. As the radiology AI ecosystem continues to grow and mature, a formalization of system assessment and evaluation is paramount to ensure patient safety, relevance and support to clinical workflows, and optimal allocation of limited AI development and validation resources before broader implementation into clinical practice. To fulfil these needs, we provide a glossary for AI software types, use cases and roles within the clinical workflow; list healthcare needs, key performance indicators and required information about software prior to assessment; and lay out examples of software performance metrics per software category. This conceptual framework is intended to streamline communication with the AI software industry and provide healthcare decision makers and radiologists with tools to assess the potential use of these software. The proposed software evaluation framework lays the foundation for a radiologist-led prospective validation network of radiology AI software. Learning Points: The rapid expansion of AI applications in radiology requires standardization of AI software specification, classification, and evaluation. The Canadian Association of Radiologists' AI Tech & Apps Working Group Proposes an AI Specification document format and supports the implementation of a clinical expert evaluation process for Radiology AI software.

Cystic fibrosis in the 21st century: what every radiologist should know.

Cystic Fibrosis (CF) is the most common lethal genetic disorder in Caucasian populations, affecting roughly 70,000 individuals worldwide. This autosomal recessive disorder causes a wide spectrum of multisystemic manifestations, most of which are either directly or indirectly related to defective epithelial chloride secretion. The current median life expectancy is 44 years; however, a significant proportion of the CF population now live into the 5th decade and beyond due to advances in treatment. As life expectancy of CF patients increases, there is a newly emerging adult CF population with unique radiological manifestations spanning multiple organ systems, which often require follow-up imaging. The goal of this article is to review the multiple systemic manifestations and complications of CF on different imaging modalities and explore the appropriate radiological follow up recommended.

Derivation and Validation of a Diagnostic Prediction Tool for Interstitial Lung Disease.

BACKGROUND: Interstitial lung disease (ILD) results in high morbidity and health-care utilization. Diagnostic delays remain common and often occur in nonpulmonology settings. Screening for ILD in these settings has the potential to reduce diagnostic delays and improve patient outcomes. RESEARCH QUESTION: This study sought to determine whether a pulmonary function test (PFT)-derived diagnostic prediction tool (ILD-Screen) could accurately identify incident ILD cases in patients undergoing PFT in nonpulmonology settings. STUDY DESIGN AND METHODS: Clinical and physiologic PFT variables predictive of ILD were identified by using iterative multivariable logistic regression models. ILD status was determined by using a multi-reader approach. An ILD-Screen score was generated by using final regression model coefficients, with a score ≥ 8 considered positive. ILD-Screen test performance was validated in an independent external cohort and applied prospectively to PFTs over 1 year to identify incident ILD cases at our institution. RESULTS: Variables comprising the ILD-Screen were age, height, total lung capacity, FEV1, diffusion capacity, and PFT indication. The ILD-Screen showed consistent test performance across cohorts, with a sensitivity of 0.79 and a specificity of 0.83 when applied prospectively. A positive ILD-Screen strongly predicted ILD (OR, 18.6; 95% CI, 9.4-36.9) and outperformed common ILD clinical features, including cough, dyspnea, lung crackles, and restrictive lung physiology. Prospective ILD-Screen application resulted in a higher proportion of patients undergoing chest CT imaging compared with a historical control cohort (74% vs 56%, respectively; P = .003), with a significantly shorter median time to chest CT imaging (5.6 vs 21.1 months; P < .001). INTERPRETATION: The ILD-Screen showed good test performance in predicting ILD across diverse geographic settings and when applied prospectively. Systematic ILD-Screen application has the potential to reduce diagnostic delays and facilitate earlier intervention in patients with ILD.

Successful lung transplantation in an HIV seropositive patient with desquamative interstitial pneumonia: a case report.

BACKGROUND: Until recently, lung transplantation was not considered in patients with human immunodeficiency virus (HIV). HIV seropositive patients with suppressed viral loads can now expect long-term survival with the advent of highly active antiretroviral therapies (HAART); however, HIV remains a relative contraindication to lung transplantation. We describe, to our knowledge, the first HIV seropositive lung transplant recipient in Canada. We also review the literature of previously reported cases of solid-organ transplantation in patients with HIV with a focus on immunosuppression considerations. CASE PRESENTATION: A 48-year old man received a bilateral lung transplant for a diagnosis of desquamative interstitial pneumonia (DIP) attributed to cigarette and cannabis smoking. His control of HIV infection pre-transplant was excellent on HAART, and he had no other contraindications to lung transplantation. The patient underwent bilateral lung transplantation using basiliximab, methylprednisolone, and mycophenolate mofetil (MMF) as induction immunosuppression. He was maintained on MMF, prednisone, and tacrolimus thereafter, and restarted his HAART regimen immediately post-operatively. His post-transplant course was complicated by Grade A1 minimal acute cellular rejection, as well as an enterovirus/rhinovirus graft infection. Despite these complications, his functional status and control of HIV infection remain excellent 24 months post-transplant. CONCLUSIONS: Our patient is one of only several HIV seropositive lung transplant recipients reported globally. With growing acceptance of transplantation in this population, there is a need for clarification of prognosis post-transplantation, as well as optimal immunosuppression regimens for these patients. This case report adds to the recent literature that suggests HIV seropositivity should not be considered a contraindication to lung transplantation, and that post-transplant patients with HIV can be managed safely with basiliximab, tacrolimus, MMF and prednisone.

Severe radiation pneumonitis after lung stereotactic ablative radiation therapy in patients with interstitial lung disease.

PURPOSE: To investigate the incidence and predictive factors of severe radiation pneumonitis (RP) after stereotactic ablative radiation therapy (SABR) in early-stage lung cancer patients with preexisting radiological interstitial lung disease (ILD). METHODS AND MATERIALS: A retrospective analysis of patients with stage I lung cancer treated with SABR from 2009 to 2014 was conducted. Interstitial lung disease diagnosis and grading was based on pretreatment high-resolution computed tomography imaging. A central review of pretreatment computed tomography by a single experienced thoracic radiologist was conducted. Univariate and multivariate analyses were conducted to determine potential predictors of severe RP in patients with ILD. RESULTS: Among 504 patients treated with SABR in this period, 6% were identified as having preexisting ILD. There was a 4% rate of ≥ grade 3 RP in the entire cohort. Interstitial lung disease was associated with increased risk of ≥ grade 3 RP (32% in ILD+ vs 2% in ILD-, P < .001). Five patients (21%) with ILD developed grade 5 RP. Lower forced expiratory volume in 1 second and forced vital capacity, higher V5Gy and mean lung dose, presence of severe radiological ILD, and combined emphysema were significant predictors of ≥ grade 3 RP on univariate analysis; only forced expiratory volume in 1 second remained on multivariate analysis. CONCLUSION: Interstitial lung disease is associated with an increased risk of severe RP after SABR. Chest imaging should be reviewed for ILD before SABR, and the risk of fatal RP should be carefully weighed against the benefits of SABR in this subgroup.

Percutaneous CT-guided lung interventions-local pleural anesthesia.

Local pleural anesthesia during percutaneous lung interventions is an important part of the procedure that has not been standardized in the literature. Significant pain can be experienced during lung biopsies, which may affect the outcome of intervention. We describe a step-by-step method of anesthesia targeting the most sensitive anatomical structure involved: the parietal pleura.

Excellent Cancer Outcomes Following Patient-adapted Robotic Lung SBRT But a Case for Caution in Idiopathic Pulmonary Fibrosis.

The aim of this study is to report outcomes and prognostic factors for early stage non-small cell lung cancer treated with patient-adapted Cyberknife stereotactic body radiotherapy. A retrospective analysis of 150 patients with T1-2N0 non-small cell lung cancer treated with stereotactic body radiotherapy was conducted. An algorithm based on tumor and patient's characteristics was used to orient patients towards soft tissue (Xsight Lung), fiducials or adjacent bone (Xsight Spine) tracking. Median biological effective dose without correction for tissue inhomogeneities was 180 Gy10 for peripheral tumors and 113 Gy10 for central tumors. Median follow-up was 22 months. Actuarial 2 years local control, overall survival and disease-specific survival were respectively 96%, 87% and 95%. Every 1 cm increase in tumor diameter was associated with a relative risk for regional or distant relapse of 2 (95%CI = 1.2-3.6, p = 0.009). With doses ≥132 Gy10 and <132 Gy10, local control was 98% vs. 82% (p = 0.07), disease-specific survival 97% vs. 78% (p = 0.02) and overall survival 93% vs. 76% (p = 0.01), respectively. Better disease-specific survival and a trend for better overall survival was observed for peripheral vs. central tumors (96% vs. 79%, p = 0.05 and 92% vs. 74%, p = 0.08, respectively). A higher Charlson comordibity score (≥4) predicted lower overall survival (79% vs. 98%, p = 0.01). Toxicities included 3 patients with idiopathic pulmonary fibrosis who developed grade 5 pneumonitis and 2 patients with grade 3 pneumonitis. We therefore report excellent local control and disease-specific survival following patient-adapted Cyberknife lung stereotactic body radiotherapy. Although toxicities were in general minimal, patients with pulmonary fibrosis might be at greater risk of severe complications. Small size, peripheral location, dose ≥ 132 Gy10 and a low Charlson co-morbidity score seem to be associated with better outcomes.

Postoperative imaging after lung transplantation.

Lung transplantation (LT) is an established procedure for chronic end-stage lung diseases. Complications are frequent and diverse and are the consequence of the complex surgical technique, the severity of the initial pathology, and the deep state of posttransplantation immunosuppression. Complications following LT include primary graft dysfunction, rejection (hyperacute, acute, and chronic), infections, posttransplantation lymphoproliferative disease, pleural and airway complications, native lung complications, and recurrence of primary disease. An understanding of these complications, their temporal evolution, and the role of radiology and other diagnostic methods in their diagnosis and management will help reduce the morbidity and mortality associated with LT.

Pleural mesothelioma surveillance: validity of cases from a tumour registry.

BACKGROUND: Pleural mesothelioma is a rare tumour associated with exposure to asbestos fibres. Fewer than than one-quarter of cases registered in the Quebec Tumour Registry (QTR) have been compensated as work-related. While establishing a surveillance system, this led to questioning as to whether there has been over-registration of cases that are not authentic pleural mesotheliomas in the QTR. OBJECTIVE: To assess whether registered cases of pleural mesothelioma could be confirmed. METHODS: A medical chart review was designed to assess the proportion of mesothelioma cases newly registered in the QTR in 2001/2002 that could be confirmed. For each registered case, clinical, medical imaging and pathology information were sought and, occasionally, additional immunohistochemistry staining was obtained. Three specialists - a chest physician, a radiologist and a pathologist - reviewed the available information and material, coding each mesothelioma case as to degree of certainty of the mesothelioma diagnosis. RESULTS: The QTR reported 190 incident cases of mesothelioma (81% males) for the period. The specialists classified 81% of charts as 'certain/probable' or 'possible' mesotheliomas, 8% as 'unlikely to be a mesothelioma' and 11% as 'not a mesothelioma'. After excluding chart summaries of unsatisfactory quality, 87% to 88% of the charts were classified as 'certain/probable' or 'possible' mesotheliomas, and 9% to 11% were still considered 'not a mesothelioma'. CONCLUSION: Tumour registry data are a valid source of information for mesothelioma surveillance. While there is some over-registration of mesothelioma cases in the QTR, a significant majority of registered cases appeared to be authentic. Over-registration cannot explain the greater proportion of cases that were not compensated.

Pandemic influenza A (H1N1) 2009: chest radiographic findings from 147 proven cases in the Montreal area.

OBJECTIVE: To describe chest radiographic findings in patients with isolated and complicated acute novel influenza A (H1N1) virus infection. METHODS: Retrospective study of 147 patients (64 men, mean age 41) with reverse-transcriptase polymerase chain reaction confirmed acute influenza A (H1N1) infection, who also had a chest radiograph <72 hours of viral specimen collection. Radiographs were analysed for acute findings. A correlation with bacterial cultures results was performed. The unpaired 2-sample equal-variance Student t test was applied to continuous variables and the Pearson χ(2) test of association to discrete variables. RESULTS: In 71% of cases, chest radiograph was normal. The presence of acute imaging findings was associated with older age (P < .05), increased number of comorbidities (most commonly, chronic obstructive pulmonary disease, diabetes, asthma) (P < .05), higher rate of hospitalization (P < .05) and intensive care unit admission, and increased mortality. Predominant acute radiographic finding in isolated influenza A (H1N1) was alveolar opacity (88%), either unifocal or multifocal, most often in the lower lobes. In the subgroup of patients with positive imaging findings and for whom nonviral microbiologic data was available, 62% had superimposed bacterial or fungal infection. CONCLUSION: In the majority of patients with acute influenza A (H1N1) infection, the chest radiograph is normal. Acute imaging findings are associated with older age, an increased number of comorbidities, and a higher rate of complications and mortality. The predominant radiographic finding of isolated primary influenza A (H1N1) infection is alveolar opacity. Superimposed bacterial infection is frequent and must be excluded in patients with abnormal imaging.