Towards texture accurate slice interpolation of medical images using PixelMiner.

Quantitative image analysis models are used for medical imaging tasks such as registration, classification, object detection, and segmentation. For these models to be capable of making accurate predictions, they need valid and precise information. We propose PixelMiner, a convolution-based deep-learning model for interpolating computed tomography (CT) imaging slices. PixelMiner was designed to produce texture-accurate slice interpolations by trading off pixel accuracy for texture accuracy. PixelMiner was trained on a dataset of 7829 CT scans and validated using an external dataset. We demonstrated the model's effectiveness by using the structural similarity index (SSIM), peak signal to noise ratio (PSNR), and the root mean squared error (RMSE) of extracted texture features. Additionally, we developed and used a new metric, the mean squared mapped feature error (MSMFE). The performance of PixelMiner was compared to four other interpolation methods: (tri-)linear, (tri-)cubic, windowed sinc (WS), and nearest neighbor (NN). PixelMiner produced texture with a significantly lowest average texture error compared to all other methods with a normalized root mean squared error (NRMSE) of 0.11 (p < .01), and the significantly highest reproducibility with a concordance correlation coefficient (CCC) ≥ 0.85 (p < .01). PixelMiner was not only shown to better preserve features but was also validated using an ablation study by removing auto-regression from the model and was shown to improve segmentations on interpolated slices.

Favorable long term effects of intensified immunosuppression combined with therapeutic plasma exchange in patients with early-onset progressive systemic sclerosis-related interstitial lung disease.

Objective: Systemic sclerosis (SSc) related mortality and morbidity remains high. Immunosuppressive therapy is considered most effective when immune activity and inflammation but not fibrosis still dominates the disease process. This study evaluated long-term intensified immunosuppression combined with therapeutic plasma exchange (TPE) in early-onset progressive SSc-related interstitial lung disease (ILD). Methods: The study cohort consisted of 161 SSc patients, with a median follow-up time of 8.9 years. The standardized mortality rate (SMR) and overall survival was calculated in patients with and without cardiopulmonary involvement. We used a standardized, pragmatic, non-randomized approach to treat 24 consecutive early progressive SSc-ILD patients with intensified immunosuppressive therapy, including plasma exchange. Outcome measurements were event-free survival (EFS), pulmonary function and safety profile. The outcome was compared with the analyzed data from the other SSc-ILD patients, who did not fulfill the inclusion criteria, and instead were treated with estimated optimal care (EOc). Results: The age-adjusted SMR of all 161 SSc patients was 3.0 (CI95%; 0.32-5.68). EFS at 10 years was 49.9% in the intensified treatment group and 43.3% in the EOc group (p = 0.106). Improvement of the percentage of predicted forced vital capacity (%pFVC) and percentage of predicted diffusing capacity for carbon monoxide (%pDLco) in the intensified treatment group was +10.1% respectively +3.6%, compared to a decrease of respectively 10.8% and 7% in the EOc (p < 0.001 resp. p = 0.019). Safety analysis showed 1 death (female patient, over 75 years of age), due to pneumosepsis, in the intensified treatment group. Conclusion: Intensified and long-lasting immunosuppression combined with TPE is safe in early severe systemic sclerosis and is associated with improved EFS and pulmonary function as compared to the outcome in the variable but EOc group. Our findings warrant larger studies for confirmation.

Yield of Adding chest CT to Abdominal CT to Detect COVID-19 in Patients Presenting With Acute Gastrointestinal Symptoms (SCOUT-3): Multicenter Study.

OBJECTIVE: To determine the incremental yield of standardized addition of chest CT to abdominal CT to detect COVID-19 in patients presenting with primarily acute gastrointestinal symptoms requiring abdominal imaging. Summary Background Data: Around 20% of patients with COVID-19 present with gastrointestinal symptoms. COVID-19 might be neglected in these patients, as the focus could be on finding abdominal pathology. During the COVID-19 pandemic, several centers have routinely added chest CT to abdominal CT to detect possible COVID-19 in patients presenting with gastrointestinal symptoms. However, the incremental yield of this strategy is unknown. METHODS: This multicenter study in 6 Dutch centers included consecutive adult patients presenting with acute nontraumatic gastrointestinal symptoms, who underwent standardized combined abdominal and chest CT between March 15, 2020 and April 30, 2020. All CT scans were read for signs of COVID-19 related pulmonary sequelae using the СО-RADS score. The primary outcome was the yield of high COVID-19 suspicion (СО-RADS 4-5) based on chest CT. RESULTS: A total of 392 patients were included. Radiologic suspicion for COVID-19 (СО-RADS 4-5) was present in 17 (4.3%) patients, eleven of which were diagnosed with COVID-19. Only 5 patients with СО-RADS 4-5 presented without any respiratory symptoms and were diagnosed with COVID-19. No relation with community prevalence could be detected. CONCLUSION: The yield of adding chest CT to abdominal CT to detect COVID-19 in patients presenting with acute gastrointestinal symptoms is extremely low with an additional detection rate of around 1%.

Exploring imaging features of molecular subtypes of large cell neuroendocrine carcinoma (LCNEC).

OBJECTIVES: Radiological characteristics and radiomics signatures can aid in differentiation between small cell lung carcinoma (SCLC) and non-small cell lung carcinoma (NSCLC). We investigated whether molecular subtypes of large cell neuroendocrine carcinoma (LCNEC), i.e. SCLC-like (with pRb loss) vs. NSCLC-like (with pRb expression), can be distinguished by imaging based on (1) imaging interpretation, (2) semantic features, and/or (3) a radiomics signature, designed to differentiate between SCLC and NSCLC. MATERIALS AND METHODS: Pulmonary oncologists and chest radiologists assessed chest CT-scans of 44 LCNEC patients for 'small cell-like' or 'non-small cell-like' appearance. The radiologists also scored semantic features of 50 LCNEC scans. Finally, a radiomics signature was trained on a dataset containing 48 SCLC and 76 NSCLC scans and validated on an external set of 58 SCLC and 40 NSCLC scans. This signature was applied on scans of 28 SCLC-like and 8 NSCLC-like LCNEC patients. RESULTS: Pulmonary oncologists and radiologists were unable to differentiate between molecular subtypes of LCNEC and no significant differences in semantic features were found. The area under the receiver operating characteristics curve of the radiomics signature in the validation set (SCLC vs. NSCLC) was 0.84 (95% confidence interval (CI) 0.77-0.92) and 0.58 (95% CI 0.29-0.86) in the LCNEC dataset (SCLC-like vs. NSCLC-like). CONCLUSION: LCNEC appears to have radiological characteristics of both SCLC and NSCLC, irrespective of pRb loss, compatible with the SCLC-like subtype. Imaging interpretation, semantic features and our radiomics signature designed to differentiate between SCLC and NSCLC were unable to separate molecular LCNEC subtypes, which underscores that LCNEC is a unique disease.

Quantitative computed tomography in COPD: possibilities and limitations.

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease that is characterized by chronic airflow limitation. Unraveling of this heterogeneity is challenging but important, because it might enable more accurate diagnosis and treatment. Because spirometry cannot distinguish between the different contributing pathways of airflow limitation, and visual scoring is time-consuming and prone to observer variability, other techniques are sought to start this phenotyping process. Quantitative computed tomography (CT) is a promising technique, because current CT technology is able to quantify emphysema, air trapping, and large airway wall dimensions. This review focuses on CT quantification techniques of COPD disease components and their current status and role in phenotyping COPD.

The relationship between lung function impairment and quantitative computed tomography in chronic obstructive pulmonary disease.

OBJECTIVES: To determine the relationship between lung function impairment and quantitative computed tomography (CT) measurements of air trapping and emphysema in a population of current and former heavy smokers with and without airflow limitation. METHODS: In 248 subjects (50 normal smokers; 50 mild obstruction; 50 moderate obstruction; 50 severe obstruction; 48 very severe obstruction) CT emphysema and CT air trapping were quantified on paired inspiratory and end-expiratory CT examinations using several available quantification methods. CT measurements were related to lung function (FEV(1), FEV(1)/FVC, RV/TLC, Kco) by univariate and multivariate linear regression analysis. RESULTS: Quantitative CT measurements of emphysema and air trapping were strongly correlated to airflow limitation (univariate r-squared up to 0.72, p < 0.001). In multivariate analysis, the combination of CT emphysema and CT air trapping explained 68-83% of the variability in airflow limitation in subjects covering the total range of airflow limitation (p < 0.001). CONCLUSIONS: The combination of quantitative CT air trapping and emphysema measurements is strongly associated with lung function impairment in current and former heavy smokers with a wide range of airflow limitation.

Screening for emphysema via exhaled volatile organic compounds.

Chronic obstructive pulmonary disease (COPD)/emphysema risk groups are well defined and screening allows for early identification of disease. The capability of exhaled volatile organic compounds (VOCs) to detect emphysema, as found by computed tomography (CT) in current and former heavy smokers participating in a lung cancer screening trial, was investigated. CT scans, pulmonary function tests and breath sample collections were obtained from 204 subjects. Breath samples were analyzed with a proton-transfer reaction mass spectrometer (PTR-MS) to obtain VOC profiles listed as ions at various mass-to-charge ratios (m/z). Using bootstrapped stepwise forward logistic regression, we identified specific breath profiles as a potential tool for the diagnosis of emphysema, of airflow limitation or gas-exchange impairment. A marker for emphysema was found at m/z 87 (tentatively attributed to 2-methylbutanal). The area under the receiver operating characteristic curve (ROC) of this marker to diagnose emphysema was 0.588 (95% CI 0.453-0.662). Mass-to-charge ratios m/z 52 (most likely chloramine) and m/z 135 (alkyl benzene) were linked to obstructive disease and m/z 122 (most probably alkyl homologs) to an impaired diffusion capacity. ROC areas were 0.646 (95% CI 0.562-0.730) and 0.671 (95% CI 0.524-0.710), respectively. In the screening setting, exhaled VOCs measured by PTR-MS constitute weak markers for emphysema, pulmonary obstruction and impaired diffusion capacity.

A large-scale evaluation of automatic pulmonary nodule detection in chest CT using local image features and k-nearest-neighbour classification.

A scheme for the automatic detection of nodules in thoracic computed tomography scans is presented and extensively evaluated. The algorithm uses the local image features of shape index and curvedness in order to detect candidate structures in the lung volume and applies two successive k-nearest-neighbour classifiers in the reduction of false-positives. The nodule detection system is trained and tested on three databases extracted from a large-scale experimental screening study. The databases are constructed in order to evaluate the algorithm on both randomly chosen screening data as well as data containing higher proportions of nodules requiring follow-up. The system results are extensively evaluated including performance measurements on specific nodule types and sizes within the databases and on lesions which later proved to be malignant. In a random selection of 813 scans from the screening study a sensitivity of 80% with an average 4.2 false-positives per scan is achieved. The detection results presented are a realistic measure of a CAD system performance in a low-dose screening study which includes a diverse array of nodules of many varying sizes, types and textures.

Nitric oxide diffusing capacity versus spirometry in the early diagnosis of emphysema in smokers.

The diffusion capacity for nitric oxide (DLNO) is independent of pulmonary capillary blood volume and equals the membrane diffusing capacity. Therefore the DLNO could be more sensitive in detecting alveolar destruction than the DLCO. We measured flow-volumes curves, DLNO, DLCO, the transfer coefficients KNO (DLNO/VA) and KCO (DLCO/VA) and performed computed tomography (CT) scans in 263 randomly selected heavy smokers. Subjects with areas > or =1% of the total lung volume showing an attenuation <-950 Hounsfield Units were considered to have emphysema. In 36 subjects emphysema was diagnosed with CT, a low KNO was present in 94 subjects, and in 95 subjects a FEV1/FVC ratio <70% was seen. The area under the ROC curve for detection CT-based emphysema was 0.894 for the KNO, 0.822 for the KCO and 0.795 for FEV1/FVC, meaning that the KNO has a slightly higher sensitivity to detect emphysema than the KCO and FEV1/FVC. The positive predictive value of KNO however was low (34.7%), while the negative predictive value of KNO was very high (98.2%), indicating an emphysema exclusion test. The DLNO/DLCO ratio is significantly higher in the study group compared to normal subjects.

Accuracy of automated volumetry of pulmonary nodules across different multislice CT scanners.

The purpose of this study was to compare the accuracy of an automated volumetry software for phantom pulmonary nodules across various 16-slice multislice spiral CT (MSCT) scanners from different vendors. A lung phantom containing five different nodule categories (intraparenchymal, around a vessel, vessel attached, pleural, and attached to the pleura), with each category comprised of 7-9 nodules (total, n = 40) of varying sizes (diameter 3-10 mm; volume 6.62 mm(3)-525 mm(3)), was scanned with four different 16-slice MSCT scanners (Siemens, GE, Philips, Toshiba). Routine and low-dose chest protocols with thin and thick collimations were applied. The data from all scanners were used for further analysis using a dedicated prototype volumetry software. Absolute percentage volume errors (APE) were calculated and compared. The mean APE for all nodules was 8.4% (+/-7.7%) for data acquired with the 16-slice Siemens scanner, 14.3% (+/-11.1%) for the GE scanner, 9.7% (+/-9.6%) for the Philips scanner and 7.5% (+/-7.2%) for the Toshiba scanner, respectively. The lowest APEs were found within the diameter size range of 5-10 mm and volumes >66 mm(3). Nodule volumetry is accurate with a reasonable volume error in data from different scanner vendors. This may have an important impact for intraindividual follow-up studies.

Sentinel lymph node investigation in melanoma: detailed analysis of the yield from step sectioning and immunohistochemistry.

AIMS: To evaluate in detail the extent to which step sectioning and immunohistochemical examination of sentinel lymph nodes (SLNs) in patients with melanoma reveal additional node positive patients, to arrive at a sensitive yet workable protocol for histopathological SLN examination. METHODS: The study comprised 29 patients with one or more positive SLN after a successful SLN procedure for clinical stage I/II melanoma. SLNs were lamellated into pieces of approximately 0.5 cm in size. One initial haematoxylin and eosin (H&E) stained central cross section was made for each block. When negative, four step ribbons were cut at intervals of 250 microm. One section from each ribbon was stained with H&E, and one was used for immunohistochemistry (IHC). RESULTS: When taking the cumulative total of detected metastases at level 5 as 100%, the percentage of SLN positive patients increased from 79%, 83%, 83%, 90% to 93% in the H&E sections through levels 1-5, and with IHC these values were 83%, 86%, 90%, 97%, and 100%, respectively. One of six patients in whom metastases were detected at levels 2-5 only had metastases in the subsequent additional lymph node dissection. CONCLUSIONS: Multiple level sectioning of SLNs (five levels at 250 microm intervals) and the use of IHC detects additional metastases up to the last level in melanoma SLNs. Although more levels of sectioning might increase the yield even further, this protocol ensures a reasonable workload for the pathologist with an acceptable sensitivity when compared with the published literature.

Clinical outcome of stage I/II melanoma patients after selective sentinel lymph node dissection: long-term follow-up results.

PURPOSE: Although sentinel lymph node (SLN) status is part of the new American Joint Committee on Cancer staging system, there is no final proof that the SLN procedure in melanoma patients influences outcome of disease. This study investigated the accuracy of the SLN procedure and clinical outcome in melanoma patients after at least 60 months of follow-up. PATIENTS AND METHODS: Between 1993 and 1996, 209 patients with stage I/II cutaneous melanoma underwent selective SLN dissection by the triple technique. If the SLN contained metastatic disease, a completion lymphadenectomy was performed. Survival analyses were performed using the Kaplan-Meier approach. Factors associated with survival were analyzed using the Cox proportional hazards regression model. RESULTS: The success rate was 99.5%. Median follow-up was 72 months. Forty patients (19%) had a positive SLN. The false-negative rate was 9%. Five-year overall survival was 87% for the entire group and 92% and 67% for SLN-negative and SLN-positive patients (P <.0001), respectively. All patients with a positive SLN and a Breslow thickness < or = 1.00 mm survived, and SLN-positive patients with a Breslow thickness less than 2.00 mm tend to have a better prognosis compared with SLN-negative patients with a Breslow thickness greater than 2.00 mm. SLN status (P =.002), Breslow thickness (P =.002), and lymphatic invasion (P =.0009) were all found to be independent prognostic factors for overall survival. CONCLUSION: With a success rate of 99.5% and a false-negative rate of 9% after long-term follow-up, the triple-technique SLN procedure is a reliable and accurate method. Survival data seem promising, although a therapeutic effect is still questionable. As shown in this study, not all SLN-positive patients have a poor prognosis.