The aim of this study was to assess the effect of applying ACR Lung-RADS in a clinical CT lung screening program on the frequency of positive and false-negative findings.
Consecutive, clinical CT ...lung screening examinations performed from January 2012 through May 2014 were retroactively reclassified using the new ACR Lung-RADS structured reporting system. All examinations had initially been interpreted by radiologists credentialed in structured CT lung screening reporting following the National Comprehensive Cancer Network’s Clinical Practice Guidelines in Oncology: Lung Cancer Screening (version 1.2012), which incorporated positive thresholds modeled after those in the National Lung Screening Trial. The positive rate, number of false-negative findings, and positive predictive value were recalculated using the ACR Lung-RADS-specific positive solid/part-solid nodule diameter threshold of 6 mm and nonsolid (ground-glass) threshold of 2 cm. False negatives were defined as cases reclassified as benign under ACR Lung-RADS that were diagnosed with malignancies within 12 months of the baseline examination.
A total of 2,180 high-risk patients underwent baseline CT lung screening during the study interval; no clinical follow-up was available in 577 patients (26%). ACR Lung-RADS reduced the overall positive rate from 27.6% to 10.6%. No false negatives were present in the 152 patients with >12-month follow-up reclassified as benign. Applying ACR Lung-RADS increased the positive predictive value for diagnosed malignancy in 1,603 patients with follow-up from 6.9% to 17.3%.
The application of ACR Lung-RADS increased the positive predictive value in our CT lung screening cohort by a factor of 2.5, to 17.3%, without increasing the number of examinations with false-negative results.
To assess key trends, strengths, and gaps in validation studies of the Food and Drug Administration (FDA)-regulated imaging-based artificial intelligence/machine learning (AI/ML) algorithms.
We ...audited publicly available details of regulated AI/ML algorithms in imaging from 2008 until April 2021. We reviewed 127 regulated software (118 AI/ML) to classify information related to their parent company, subspecialty, body area and specific anatomy type, imaging modality, date of FDA clearance, indications for use, target pathology (such as trauma) and findings (such as fracture), technique (CAD triage, CAD detection and/or characterization, CAD acquisition or improvement, and image processing/quantification), product performance, presence, type, strength and availability of clinical validation data. Pertaining to validation data, where available, we recorded the number of patients or studies included, sensitivity, specificity, accuracy, and/or receiver operating characteristic area under the curve, along with information on ground-truthing of use-cases. Data were analyzed with pivot tables and charts for descriptive statistics and trends.
We noted an increasing number of FDA-regulated AI/ML from 2008 to 2021. Seventeen (17/118) regulated AI/ML algorithms posted no validation claims or data. Just 9/118 reviewed AI/ML algorithms had a validation dataset sizes of over 1000 patients. The most common type of AI/ML included image processing/quantification (IPQ; n = 59/118), and triage (CADt; n = 27/118). Brain, breast, and lungs dominated the targeted body regions of interest.
Insufficient public information on validation datasets in several FDA-regulated AI/ML algorithms makes it difficult to justify clinical applications since their generalizability and presence of bias cannot be inferred.
Purpose:
Metal objects present in x-ray computed tomography (CT) scans are accompanied by physical phenomena that render CT projections inconsistent with the linear assumption made for analytical ...reconstruction. The inconsistencies create artifacts in reconstructed images. Metal artifact reduction algorithms replace the inconsistent projection data passing through metals with estimates of the true underlying projection data, but when the data estimates are inaccurate, secondary artifacts are generated. The secondary artifacts may be as unacceptable as the original metal artifacts; therefore, better projection data estimation is critical. This research uses computer vision techniques to create better estimates of the underlying projection data using observations about the appearance and nature of metal artifacts.
Methods:
The authors developed a method of estimating underlying projection data through the use of an intermediate image, called the prior image. This method generates the prior image by segmenting regions of the originally reconstructed image, and discriminating between regions that are likely to be metal artifacts and those that are likely to represent anatomical structures. Regions identified as metal artifact are replaced with a constant soft-tissue value, while structures such as bone or air pockets are preserved. This prior image is reprojected (forward projected), and the reprojections guide the estimation of the underlying projection data using previously published interpolation techniques. The algorithm is tested on head CT test cases containing metal implants and compared against existing methods.
Results:
Using the new method of prior image generation on test images, metal artifacts were eliminated or reduced and fewer secondary artifacts were present than with previous methods. The results apply even in the case of multiple metal objects, which is a challenging problem. The authors did not observe secondary artifacts that were comparable to or worse than the original metal artifacts, as sometimes occurred with the other methods. The accuracy of the prior was found to be more critical than the particular interpolation method.
Conclusions:
Metals produce predictable artifacts in CT images of the head. Using the new method, metal artifacts can be discriminated from anatomy, and the discrimination can be used to reduce metal artifacts.
The purpose of our study was to determine the clinical and CT predictors of recurrent disease after a first episode of diverticulitis that was successfully managed nonoperatively.
We retrospectively ...analyzed 954 consecutive patients who presented to our institution with diverticulitis from 2002 to 2008. Patients were identified with International Classification of Diseases, 9th Revision/Current Procedural Terminology codes. Patients were excluded if they had subsequent colectomy based on the first attack (n = 81), or if the attack they had between 2002 and 2008 was not their first attack (n = 201). We evaluated CT variables chosen by a panel of expert gastrointestinal radiologists. These radiologists reviewed the available published literature for CT imaging characteristics thought to predict diverticulitis severity. CT variables (n = 20) were determined by prospective reevaluation of scans by blinded study radiologists. Clinical variables (n = 43) were coded based on a retrospective chart review. Univariate analysis of variables in relation to recurrent disease was performed by a log-rank test of Kaplan-Meier estimates. Multivariate analysis was performed using Cox proportional hazards modeling. Variables with P < .2 on univariate analysis were included in a stepwise selection algorithm.
The study population included 672 patients; mean age, 61 ± 15 years; mean follow-up, 42.8 ± 24 months. The index presentation of diverticulitis was most commonly located in the sigmoid colon (72%), followed by descending colon (33%), right colon (5%), and transverse colon (3%). Overall recurrence at 5 years was 36% by (95% CI 31.4%-40.6%) Kaplan-Meier estimate. Complicated recurrence (fistula, abscess, free perforation) occurred in 3.9% (95% CI 2.2%-5.6%) of patients at 5 years by Kaplan-Meier estimate. Family history of diverticulitis (HR 2.2, 95% CI 1.4-3.2), length of involved colon >5 cm (HR 1.7, 95% CI 1.3-2.3), and retroperitoneal abscess (HR 4.5, 95% CI 1.1-18.4) were associated with diverticulitis recurrence. Right colon disease (HR 0.27, 95% CI 0.09-0.86) was associated with freedom from recurrence.
Although diverticulitis recurrence is common following an initial attack that has been managed medically, complicated recurrence is uncommon. Patients who present with a family history of diverticulitis, long segment of involved colon, and/or retroperitoneal abscess are at higher risk for recurrent disease. Patients who present with right-sided diverticulitis are at low risk for recurrent disease. These findings should be taken into consideration when counseling patients regarding the potential benefits of prophylactic colectomy.
Interstitial lung abnormalities (ILA) are CT findings suggestive of interstitial lung disease in individuals without a prior diagnosis or suspicion of ILD. Previous studies have demonstrated that ILA ...are associated with clinically significant outcomes including mortality. The aim of this study was to determine the prevalence of ILA in a large CT lung cancer screening program and the association with clinically significant outcomes including mortality, hospitalizations, cancer and ILD diagnosis.
This was a retrospective study of individuals enrolled in a CT lung cancer screening program from 2012 to 2014. Baseline and longitudinal CT scans were scored for ILA per Fleischner Society guidelines. The primary analyses examined the association between baseline ILA and mortality, all-cause hospitalization, and incidence of lung cancer. Kaplan-Meier plots were generated to visualize the associations between ILA and lung cancer and all-cause mortality. Cox regression proportional hazards models were used to test for this association in both univariate and multivariable models.
1699 subjects met inclusion criteria. 41 (2.4%) had ILA and 101 (5.9%) had indeterminate ILA on baseline CTs. ILD was diagnosed in 10 (24.4%) of 41 with ILA on baseline CT with a mean time from baseline CT to diagnosis of 4.47 ± 2.72 years. On multivariable modeling, the presence of ILA remained a significant predictor of death, HR 3.87 (2.07, 7.21; p < 0.001) when adjusted for age, sex, BMI, pack years and active smoking, but not of lung cancer and all-cause hospital admission. Approximately 50% with baseline ILA had progression on the longitudinal scan.
ILA identified on baseline lung cancer screening exams are associated with all-cause mortality. In addition, a significant proportion of patients with ILA are subsequently diagnosed with ILD and have CT progression on longitudinal scans.
ClinicalTrials.gov; No.: NCT04503044.
Abstract Purpose The aim of this study was to assess the effect of applying ACR Lung-RADS in a clinical CT lung screening program on the frequency of positive and false-negative findings. Methods ...Consecutive, clinical CT lung screening examinations performed from January 2012 through May 2014 were retroactively reclassified using the new ACR Lung-RADS structured reporting system. All examinations had initially been interpreted by radiologists credentialed in structured CT lung screening reporting following the National Comprehensive Cancer Network’s Clinical Practice Guidelines in Oncology: Lung Cancer Screening (version 1.2012), which incorporated positive thresholds modeled after those in the National Lung Screening Trial. The positive rate, number of false-negative findings, and positive predictive value were recalculated using the ACR Lung-RADS-specific positive solid/part-solid nodule diameter threshold of 6 mm and nonsolid (ground-glass) threshold of 2 cm. False negatives were defined as cases reclassified as benign under ACR Lung-RADS that were diagnosed with malignancies within 12 months of the baseline examination. Results A total of 2,180 high-risk patients underwent baseline CT lung screening during the study interval; no clinical follow-up was available in 577 patients (26%). ACR Lung-RADS reduced the overall positive rate from 27.6% to 10.6%. No false negatives were present in the 152 patients with >12-month follow-up reclassified as benign. Applying ACR Lung-RADS increased the positive predictive value for diagnosed malignancy in 1,603 patients with follow-up from 6.9% to 17.3%. Conclusions The application of ACR Lung-RADS increased the positive predictive value in our CT lung screening cohort by a factor of 2.5, to 17.3%, without increasing the number of examinations with false-negative results.
Artificial Intelligence (AI) carries the potential for unprecedented disruption in radiology, with possible positive and negative consequences. The integration of AI in radiology holds the potential ...to revolutionize healthcare practices by advancing diagnosis, quantification, and management of multiple medical conditions. Nevertheless, the ever-growing availability of AI tools in radiology highlights an increasing need to critically evaluate claims for its utility and to differentiate safe product offerings from potentially harmful, or fundamentally unhelpful ones.
This multi-society paper, presenting the views of Radiology Societies in the USA, Canada, Europe, Australia, and New Zealand, defines the potential practical problems and ethical issues surrounding the incorporation of AI into radiological practice. In addition to delineating the main points of concern that developers, regulators, and purchasers of AI tools should consider prior to their introduction into clinical practice, this statement also suggests methods to monitor their stability and safety in clinical use, and their suitability for possible autonomous function. This statement is intended to serve as a useful summary of the practical issues which should be considered by all parties involved in the development of radiology AI resources, and their implementation as clinical tools.
Key points
• The incorporation of artificial intelligence (AI) in radiological practice demands increased monitoring of its utility and safety.
• Cooperation between developers, clinicians, and regulators will allow all involved to address ethical issues and monitor AI performance.
• AI can fulfil its promise to advance patient well-being if all steps from development to integration in healthcare are rigorously evaluated.
Cancer therapy has evolved from being broadly directed towards tumor types, to highly specific treatment protocols that target individual molecular subtypes of tumors. With the ever-increasing data ...on imaging characteristics of tumor subtypes and advancements in imaging techniques, it is now often possible for radiologists to differentiate tumor subtypes on imaging. Armed with this knowledge, radiologists may be able to provide specific information that can obviate the need for invasive methods to identify tumor subtypes. Different tumor subtypes also differ in their patterns of metastatic spread. Awareness of these differences can direct radiologists to relevant anatomical sites to screen for early metastases that may otherwise be difficult to detect during cursory inspection. Likewise, this knowledge will help radiologists to interpret indeterminate findings in a more specific manner.
•Tumor subtypes can be identified based on their different imaging characteristics.•Awareness of tumor subtype can help radiologists chose the appropriate modality for additional imaging workup.•Awareness of differences in metastatic pattern between tumor subtypes can be helpful to identify early metastases.
Lung cancer screening with low-dose computed tomography is proven to reduce lung cancer mortality among high-risk patients. However, critics raise concern over the potential for unnecessary surgical ...procedures performed for benign disease as a result of screening. We reviewed our outcomes in a large clinical lung cancer screening program to assess the number of surgical procedures done for benign disease, as we believe this is an important quality metric.
We retrospectively reviewed our surgical outcomes of consecutive patients who underwent low-dose computed tomography lung cancer screening from January 2012 through June 2014 using a prospectively collected database. All patients met the National Comprehensive Cancer Network lung cancer screening guidelines high-risk criteria.
There were 1,654 screened patients during the study interval with clinical follow-up at Lahey Hospital & Medical Center. Twenty-five of the 1,654 (1.5%) had surgery. Five of 25 had non-lung cancer diagnoses: 2 hamartomas, 2 necrotizing granulomas, and 1 breast cancer metastasis. The incidence of surgery for non-lung cancer diagnosis was 0.30% (5 of 1,654), and the incidence of surgery for benign disease was 0.24% (4 of 1,654). Twenty of 25 had lung cancer, 18 early stage and 2 late stage. There were no surgery-related deaths, and there was 1 major surgical complication (4%) at 30 days.
The incidence of surgical intervention for non-lung cancer diagnosis was low (0.30%) and is comparable to the rate reported in the National Lung Screening Trial (0.62%). Surgical intervention for benign disease was rare (0.24%) in our experience.