Development and multicenter validation of deep convolutional neural network-based detection of colorectal cancer on abdominal CT.

OBJECTIVES: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network. METHODS: This retrospective study included consecutive patients with colorectal adenocarcinoma who underwent abdominal CT before CRC resection surgery (training set = 379, test set = 103). We customized the 3D U-Net of nnU-Net (CUNET) for CRC detection, which was trained with fivefold cross-validation using annotated CT images. CUNET was validated using datasets covering various clinical situations and institutions: an internal test set (n = 103), internal patients with CRC first determined by CT (n = 54) and asymptomatic CRC (n = 51), and an external validation set from two institutions (n = 60). During each validation, data from the healthy population were added (internal = 60; external = 130). CUNET was compared with other deep CNNs: residual U-Net and EfficientDet. The CAD performances were evaluated using per-CRC sensitivity (true positive/all CRCs), free-response receiver operating characteristic (FROC), and jackknife alternative FROC (JAFROC) curves. RESULTS: CUNET showed a higher maximum per-CRC sensitivity than residual U-Net and EfficientDet (internal test set 91.3% vs. 61.2%, and 64.1%). The per-CRC sensitivity of CUNET at false-positive rates of 3.0 was as follows: internal CRC determined by CT, 89.3%; internal asymptomatic CRC, 87.3%; and external validation, 89.6%. CUNET detected 69.2% (9/13) of CRCs missed by radiologists and 89.7% (252/281) of CRCs from all validation sets. CONCLUSIONS: CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions. KEY POINTS: • Customized 3D U-Net of nnU-Net (CUNET) can be applied to the opportunistic detection of colorectal cancer (CRC) in abdominal CT, helping radiologists detect unexpected CRC. • CUNET showed the best performance at false-positive rates ≥ 3.0, and 30.1% of false-positives were in the colorectum. CUNET detected 69.2% (9/13) of CRCs missed by radiologists and 87.3% (48/55) of asymptomatic CRCs. • CUNET detected CRCs in multiple validation sets composed of varying clinical situations and from different institutions, and CUNET detected 89.7% (252/281) of CRCs from all validation sets.

Deep learning referral suggestion and tumour discrimination using explainable artificial intelligence applied to multiparametric MRI.

OBJECTIVES: An appropriate and fast clinical referral suggestion is important for intra-axial mass-like lesions (IMLLs) in the emergency setting. We aimed to apply an interpretable deep learning (DL) system to multiparametric MRI to obtain clinical referral suggestion for IMLLs, and to validate it in the setting of nontraumatic emergency neuroradiology. METHODS: A DL system was developed in 747 patients with IMLLs ranging 30 diseases who underwent pre- and post-contrast T1-weighted (T1CE), FLAIR, and diffusion-weighted imaging (DWI). A DL system that segments IMLLs, classifies tumourous conditions, and suggests clinical referral among surgery, systematic work-up, medical treatment, and conservative treatment, was developed. The system was validated in an independent cohort of 130 emergency patients, and performance in referral suggestion and tumour discrimination was compared with that of radiologists using receiver operating characteristics curve, precision-recall curve analysis, and confusion matrices. Multiparametric interpretable visualisation of high-relevance regions from layer-wise relevance propagation overlaid on contrast-enhanced T1WI and DWI was analysed. RESULTS: The DL system provided correct referral suggestions in 94 of 130 patients (72.3%) and performed comparably to radiologists (accuracy 72.6%, McNemar test; p = .942). For distinguishing tumours from non-tumourous conditions, the DL system (AUC, 0.90 and AUPRC, 0.94) performed similarly to human readers (AUC, 0.81~0.92, and AUPRC, 0.88~0.95). Solid portions of tumours showed a high overlap of relevance, but non-tumours did not (Dice coefficient 0.77 vs. 0.33, p < .001), demonstrating the DL's decision. CONCLUSIONS: Our DL system could appropriately triage patients using multiparametric MRI and provide interpretability through multiparametric heatmaps, and may thereby aid neuroradiologic diagnoses in emergency settings. CLINICAL RELEVANCE STATEMENT: Our AI triages patients with raw MRI images to clinical referral pathways in brain intra-axial mass-like lesions. We demonstrate that the decision is based on the relative relevance between contrast-enhanced T1-weighted and diffusion-weighted images, providing explainability across multiparametric MRI data. KEY POINTS: • A deep learning (DL) system using multiparametric MRI suggested clinical referral to patients with intra-axial mass-like lesions (IMLLs) similar to radiologists (accuracy 72.3% vs. 72.6%). • In the differentiation of tumourous and non-tumourous conditions, the DL system (AUC, 0.90) performed similar with radiologists (AUC, 0.81-0.92). • The DL's decision basis for differentiating tumours from non-tumours can be quantified using multiparametric heatmaps obtained via the layer-wise relevance propagation method.

Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imaging.

BACKGROUND: Deep learning (DL) algorithms are playing an increasing role in automatic medical image analysis. PURPOSE: To evaluate the performance of a DL model for the automatic detection of intracranial haemorrhage and its subtypes on non-contrast CT (NCCT) head studies and to compare the effects of various preprocessing and model design implementations. METHODS: The DL algorithm was trained and externally validated on open-source, multi-centre retrospective data containing radiologist-annotated NCCT head studies. The training dataset was sourced from four research institutions across Canada, the USA and Brazil. The test dataset was sourced from a research centre in India. A convolutional neural network (CNN) was used, with its performance compared against similar models with additional implementations: (1) a recurrent neural network (RNN) attached to the CNN, (2) preprocessed CT image-windowed inputs and (3) preprocessed CT image-concatenated inputs. The area under the receiver operating characteristic curve (AUC-ROC) and microaveraged precision (mAP) score were used to evaluate and compare model performances. RESULTS: The training and test datasets contained 21,744 and 491 NCCT head studies, respectively, with 8,882 (40.8%) and 205 (41.8%) positive for intracranial haemorrhage. Implementation of preprocessing techniques and the CNN-RNN framework increased mAP from 0.77 to 0.93 and increased AUC-ROC [95% confidence intervals] from 0.854 [0.816-0.889] to 0.966 [0.951-0.980] (p-value = 3.91 × 10-12). CONCLUSIONS: The deep learning model accurately detected intracranial haemorrhage and improved in performance following specific implementation techniques, demonstrating clinical potential as a decision support tool and an automated system to improve radiologist workflow efficiency. KEY POINTS: • The deep learning model detected intracranial haemorrhages on computed tomography with high accuracy. • Image preprocessing, such as windowing, plays a large role in improving deep learning model performance. • Implementations which enable an analysis of interslice dependencies can improve deep learning model performance. • Visual saliency maps can facilitate explainable artificial intelligence systems. • Deep learning within a triage system may expedite earlier intracranial haemorrhage detection.

Classifying Temporomandibular Disorder with Artificial Intelligent Architecture Using Magnetic Resonance Imaging.

This study proposes a new diagnostic tool for automatically extracting discriminative features and detecting temporomandibular joint disc displacement (TMJDD) accurately with artificial intelligence. We analyzed the structural magnetic resonance imaging (MRI) images of 52 patients with TMJDD and 32 healthy controls. The data were split into training and test sets, and only the training sets were used for model construction. U-net was trained with 100 sagittal MRI images of the TMJ to detect the joint cavity between the temporal bone and the mandibular condyle, which was used as the region of interest, and classify the images into binary categories using four convolutional neural networks: InceptionResNetV2, InceptionV3, DenseNet169, and VGG16. The best models were InceptionV3 and DenseNet169; the results of InceptionV3 for recall, precision, accuracy, and F1 score were 1, 0.81, 0.85, and 0.9, respectively, and the corresponding results of DenseNet169 were 0.92, 0.86, 0.85, and 0.89, respectively. Automated detection of TMJDD from sagittal MRI images is a promising technique that involves using deep learning neural networks. It can be used to support clinicians in diagnosing patients as having TMJDD.

Assessment of hepatic arterial hemodynamics with 4D flow MRI: in vitro analysis of motion and spatial resolution related error and in vivo feasibility study in 20 volunteers.

OBJECTIVES: To assess the ability of four-dimensional (4D) flow MRI to measure hepatic arterial hemodynamics by determining the effects of spatial resolution and respiratory motion suppression in vitro and its applicability in vivo with comparison to two-dimensional (2D) phase-contrast MRI. METHODS: A dynamic hepatic artery phantom and 20 consecutive volunteers were scanned. The accuracies of Cartesian 4D flow sequences with k-space reordering and navigator gating at four spatial resolutions (0.5- to 1-mm isotropic) and navigator acceptance windows (± 8 to ± 2 mm) and one 2D phase-contrast sequence (0.5-mm in -plane) were assessed in vitro at 3 T. Two sequences centered on gastroduodenal and hepatic artery branches were assessed in vivo for intra - and interobserver agreement and compared to 2D phase-contrast. RESULTS: In vitro, higher spatial resolution led to a greater decrease in error than narrower navigator window (30.5 to -4.67% vs -6.64 to -4.67% for flow). In vivo, hepatic and gastroduodenal arteries were more often visualized with the higher resolution sequence (90 vs 71%). Despite similar interobserver agreement (κ = 0.660 and 0.704), the higher resolution sequence had lower variability for area (CV = 20.04 vs 30.67%), flow (CV = 34.92 vs 51.99%), and average velocity (CV = 26.47 vs 44.76%). 4D flow had lower differences between inflow and outflow at the hepatic artery bifurcation (11.03 ± 5.05% and 15.69 ± 6.14%) than 2D phase-contrast (28.77 ± 21.01%). CONCLUSION: High-resolution 4D flow can assess hepatic artery anatomy and hemodynamics with improved accuracy, greater vessel visibility, better interobserver reliability, and internal consistency. KEY POINTS: • Motion-suppressed Cartesian four-dimensional (4D) flow MRI with higher spatial resolution provides more accurate measurements even when accepted respiratory motion exceeds voxel size. • 4D flow MRI with higher spatial resolution provides substantial interobserver agreement for visualization of hepatic artery branches. • Lower peak and average velocities and a trend toward better internal consistency were observed with 4D flow MRI as compared to 2D phase-contrast.

Deep learning-based algorithm improved radiologists' performance in bone metastases detection on CT.

OBJECTIVES: To develop and evaluate a deep learning-based algorithm (DLA) for automatic detection of bone metastases on CT. METHODS: This retrospective study included CT scans acquired at a single institution between 2009 and 2019. Positive scans with bone metastases and negative scans without bone metastasis were collected to train the DLA. Another 50 positive and 50 negative scans were collected separately from the training dataset and were divided into validation and test datasets at a 2:3 ratio. The clinical efficacy of the DLA was evaluated in an observer study with board-certified radiologists. Jackknife alternative free-response receiver operating characteristic analysis was used to evaluate observer performance. RESULTS: A total of 269 positive scans including 1375 bone metastases and 463 negative scans were collected for the training dataset. The number of lesions identified in the validation and test datasets was 49 and 75, respectively. The DLA achieved a sensitivity of 89.8% (44 of 49) with 0.775 false positives per case for the validation dataset and 82.7% (62 of 75) with 0.617 false positives per case for the test dataset. With the DLA, the overall performance of nine radiologists with reference to the weighted alternative free-response receiver operating characteristic figure of merit improved from 0.746 to 0.899 (p < .001). Furthermore, the mean interpretation time per case decreased from 168 to 85 s (p = .004). CONCLUSION: With the aid of the algorithm, the overall performance of radiologists in bone metastases detection improved, and the interpretation time decreased at the same time. KEY POINTS: • A deep learning-based algorithm for automatic detection of bone metastases on CT was developed. • In the observer study, overall performance of radiologists in bone metastases detection improved significantly with the aid of the algorithm. • Radiologists' interpretation time decreased at the same time.

Artificial Intelligence in Echocardiography.

Artificial intelligence in diagnostic cardiac-imaging platforms is advancing rapidly. In particular, artificial intelligence algorithms have increased the efficiency and accuracy of echocardiographic cardiovascular imaging, resulting in more complex echocardiographic imaging techniques and expanded use among noncardiologists. Here, we provide an overview of real-world applications of artificial intelligence in echocardiography including automatic high-quality computer-optimized image acquisition sequences, automated measurements, and algorithms for the rapid and accurate interpretation of cardiac physiology. These advances will not replace physicians but will improve their productivity, workflow, and diagnostic performance.

Artificial Intelligence in Cardiovascular Medicine: Historical Overview, Current Status, and Future Directions.

Artificial intelligence and machine learning are rapidly gaining popularity in every aspect of our daily lives, and cardiovascular medicine is no exception. Here, we provide physicians with an overview of the past, present, and future of artificial intelligence applications in cardiovascular medicine. We describe essential and powerful examples of machine-learning applications in industry and elsewhere. Finally, we discuss the latest technologic advances, as well as the benefits and limitations of artificial intelligence and machine learning in cardiovascular medicine.

The impact of cardiopulmonary hemodynamic factors in volumetry for pulmonary nodule management.

BACKGROUND: The acceptance of coronary CT angiogram (CCTA) scans in the management of stable angina has led to an exponential increase in studies performed and reported incidental findings, including pulmonary nodules (PN). Using low-dose CT scans, volumetry tools are used in growth assessment and risk stratification of PN between 5 and 8 mm in diameter. Volumetry of PN could also benefit from the increased temporal resolution of CCTA scans, potentially expediting clinical decisions when an incidental PN is first detected on a CCTA scan, and allow for better resource management and planning in a Radiology department. This study aims to investigate how cardiopulmonary hemodynamic factors impact the volumetry of PN using CCTA scans. These factors include the cardiac phase, vascular distance from the main pulmonary artery (MPA) to the nodule, difference of the MPA diameter between systole and diastole, nodule location, and cardiomegaly presence. MATERIALS AND METHODS: Two readers reviewed all CCTA scans performed from 2016 to 2019 in a tertiary hospital and detected PN measuring between 5 and 8 mm in diameter. Each observer measured each nodule using two different software packages and in systole and diastole. A multiple linear regression model was applied, and inter-observer and inter-software agreement were assessed using intraclass correlation. RESULTS: A total of 195 nodules from 107 patients were included in this retrospective, cross-sectional and observational study. The regression model identified the vascular distance (p < 0.001), the difference of the MPA diameter between systole and diastole (p < 0.001), and the location within the lower or posterior thirds of the field of view (p < 0.001 each) as affecting the volume measurement. The cardiac phase was not significant in the model. There was a very high inter-observer agreement but no reasonable inter-software agreement between measurements. CONCLUSIONS: PN volumetry using CCTA scans seems to be sensitive to cardiopulmonary hemodynamic changes independently of the cardiac phase. These might also be relevant to non-gated scans, such as during PN follow-up. The cardiopulmonary hemodynamic changes are a new limiting factor to PN volumetry. In addition, when a patient experiences an acute or deteriorating cardiopulmonary disease during PN follow-up, these hemodynamic changes could affect the PN growth estimation.

DWI-related texture analysis for prostate cancer: differences in correlation with histological aggressiveness and data repeatability between peripheral and transition zones.

BACKGROUND: We investigated the correlation between texture features extracted from apparent diffusion coefficient (ADC) maps or diffusion-weighted images (DWIs), and grade group (GG) in the prostate peripheral zone (PZ) and transition zone (TZ), and assessed reliability in repeated examinations. METHODS: Patients underwent 3-T pelvic magnetic resonance imaging (MRI) before radical prostatectomy with repeated DWI using b-values of 0, 100, 1,000, and 1,500 s/mm2. Region of interest (ROI) for cancer was assigned to the first and second DWI acquisition separately. Texture features of ROIs were extracted from comma-separated values (CSV) data of ADC maps generated from several sets of two b-value combinations and DWIs, and correlation with GG, discrimination ability between GG of 1-2 versus 3-5, and data repeatability were evaluated in PZ and TZ. RESULTS: Forty-four patients with 49 prostate cancers met the eligibility criteria. In PZ, ADC 10% and 25% based on ADC map of two b-value combinations of 100 and 1,500 s/mm2 and 10% based on ADC map with b-value of 0 and 1,500 s/mm2 showed significant correlation with GG, acceptable discrimination ability, and good repeatability. In TZ, higher-order texture feature of busyness extracted from ADC map of 100 and 1,500 s/mm2, and high gray-level run emphasis, short-run high gray-level emphasis, and high gray-level zone emphasis from DWI with b-value of 100 s/mm2 demonstrated significant correlation, excellent discrimination ability, but moderate repeatability. CONCLUSIONS: Some DWI-related features showed significant correlation with GG, acceptable to excellent discrimination ability, and moderate to good data repeatability in prostate cancer, and differed between PZ and TZ.

Value of a deep learning-based algorithm for detecting Lung-RADS category 4 nodules on chest radiographs in a health checkup population: estimation of the sample size for a randomized controlled trial.

OBJECTIVE: To explore the value of a deep learning-based algorithm in detecting Lung CT Screening Reporting and Data System category 4 nodules on chest radiographs from an asymptomatic health checkup population. METHODS: Data from an annual retrospective cohort of individuals who underwent chest radiographs for health checkup purposes and chest CT scanning within 3 months were collected. Among 3073 individuals, 118 with category 4 nodules on CT were selected. A reader performance test was performed using those 118 radiographs and randomly selected 51 individuals without any nodules. Four radiologists independently evaluated the radiographs without and with the results of the algorithm; and sensitivities/specificities were compared. The sample size needed to confirm the difference in detection rates was calculated, i.e., the number of true-positive radiographs divided by the total number of radiographs. RESULTS: The sensitivity of the radiologists substantially increased aided by the algorithm (38.8% [183/472] to 45.1% [213/472]; p < .001) without significant change in specificity (94.1% [192/204] vs. 92.2% [188/204]; p = .22). Pooled radiologists detected more nodules with the algorithm (32.0% [156/488] vs. 38.9% [190/488]; p < .001), without alteration of false-positive rates (0.09 [62/676], both). Pooled detection rates for the annual cohort were 1.49% (183/12,292) and 1.73% (213/12,292) without and with the algorithm, respectively. A sample size of 41,776 in each arm would be required to demonstrate significant detection rate difference with < 5% type I error and > 80% power. CONCLUSION: Although readers substantially increased sensitivity in detecting nodules on chest radiographs from a health checkup population aided by the algorithm, detection rate difference was only 0.24%, requiring a sample size >80,000 for a randomized controlled trial. KEY POINTS: • Aided by a deep learning algorithm, pooled radiologists improved their sensitivity in detecting Lung-RADS category 4 nodules on chest radiographs from a health checkup population (38.8% [183/472] to 45.1% [213/472]; p < .001), without increasing false-positive rate. • The prevalence of the Lung-RADS category 4 nodules was 3.8% (118/3073) on the population, resulting in only 0.24% increase of the detection rate for the radiologists with assistance of the algorithm. • To confirm the significant detection rate increase by a randomized controlled trial, a sample size of 84,000 would be required.

Issues related to artificial intelligence research in ophthalmology / 中华眼底病杂志

At present, artificial intelligence (AI) has been widely used in the diagnosis and treatment of various ophthalmological diseases, but there are still many problems. Due to the lack of standardized test sets, gold standards, and recognized evaluation systems for the accuracy of AI products, it is difficult to compare the results of multiple studies. When it comes to the field of image generation, we hardly have an efficient approach to evaluating research results. In clinical practice, ophthalmological AI research is often out of touch with actual clinical needs. The requirements for the quality and quantity of clinical data put more burden on AI research, limiting the transformation of AI studies. The prediction of systemic diseases based on fundus images is making progressive advancement. However, the lack of interpretability of the research lower the acceptance. Ophthalmology AI research also suffer from ethical controversy due to unconstructed regulations and regulatory mechanisms, concerns on patients' privacy and data security, and the risk of aggravating the unfairness of medical resources.

Added value of 18F-FDG total-body PET/CT dynamic imaging in the diagnosis of liver malignant tumors / 中华核医学与分子影像杂志

Objective:To explore the added value of time-activity curve (TAC) and target-to-background ratio (TBR) obtained by 18F-FDG total-body PET/CT dynamic imaging in the diagnosis of liver malignant tumors. Methods:From December 2019 to October 2021, 109 patients (65 males, 44 females; age (59.3±9.3) years) with hepatocellular carcinoma (HCC; n=27), intrahepatic cholangiocarcinoma (ICC; n=61) and colorectal cancer with liver metastasis (CRLM; n=21) who underwent 60 min 18F-FDG total-body PET/CT dynamic imaging in Zhongshan Hospital, Fudan University were retrospectively enrolled. Dynamic PET/CT images were divided into perfusion-weighted (PW) phase and metabolism-weighted (MW) phase. The arterial phase was defined as the 15 s after the abdominal aorta peak frame at PW. TACs at MW were divided into three types as Graph A, Graph B and Graph C. One-way analysis of variance was used to compare difference of TBR 30/60 among groups. ROC curve analysis was used to evaluate diagnostic effectiveness. Results:With hypervascularity as the diagnostic standard of HCC, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 66.7%(18/27), 75.6%(59/78), 48.6%(18/37) and 86.8%(59/68), respectively. With Graph B as the diagnostic standard of HCC, the sensitivity, specificity, PPV and NPV were 44.4%(12/27), 85.4%(70/82), 50.0%(12/24) and 82.4%(70/85), respectively. The TBR 30/60 of HCC, ICC and CRLM was 0.38±0.19, 0.49±0.18 and 0.64±0.20 respectively ( F=10.89, P<0.001). When the cut-off value of TBR 30/60 was 0.43, the AUC of distinguishing HCC from ICC and CRLM was 0.72, with the sensitivity and specificity of 70.5%(55/78) and 65.2%(15/23). When the cut-off value of TBR 30/60 was 0.64, the AUC of distinguishing ICC from CRLM was 0.71, with the sensitivity and specificity of 61.9%(13/21) and 82.5%(47/57). Conclusion:TAC graph types and TBR 30/60 obtained by total-body PET/CT dynamic imaging display potential value for differentiation between hepatic tumor types.

Detecting low blood concentrations in joints using T1 and T2 mapping at 1.5, 3, and 7 T: an in vitro study.

BACKGROUND: Intra-articular blood causes irreversible joint damage, whilst clinical differentiation between haemorrhagic joint effusion and other effusions can be challenging. An accurate non-invasive method for the detection of joint bleeds is lacking. The aims of this phantom study were to investigate whether magnetic resonance imaging (MRI) T1 and T2 mapping allows for differentiation between simple and haemorrhagic joint effusion and to determine the lowest blood concentration that can be detected. METHODS: Solutions of synovial fluid with blood concentrations ranging from 0 to 100% were scanned at 1.5, 3, and 7 T. T1 maps were generated with an inversion recovery technique and T2 maps from multi spin-echo sequences. In both cases, the scan acquisition times were below 5 min. Regions of interest were manually drawn by two observers in the obtained T1 and T2 maps for each sample. The lowest detectable blood concentration was determined for all field strengths. RESULTS: At all field strengths, T1 and T2 relaxation times decreased with higher blood concentrations. The lowest detectable blood concentrations using T1 mapping were 10% at 1.5 T, 25% at 3 T, and 50% at 7 T. For T2 mapping, the detection limits were 50%, 5%, and 25%, respectively. CONCLUSIONS: T1 and T2 mapping can detect different blood concentrations in synovial fluid in vitro at clinical field strengths. Especially, T2 measurements at 3 T showed to be highly sensitive. Short acquisition times would make these methods suitable for clinical use and therefore might be promising tools for accurate discrimination between simple and haemorrhagic joint effusion in vivo.

Digital subtraction of temporally sequential mammograms for improved detection and classification of microcalcifications.

BACKGROUND: Our aim was to demonstrate that automated detection and classification of breast microcalcifications, according to Breast Imaging Reporting and Data System (BI-RADS) categorisation, can be improved with the subtraction of sequential mammograms as opposed to using the most recent image only. METHODS: One hundred pairs of mammograms were retrospectively collected from two temporally sequential rounds. Fifty percent of the images included no (BI-RADS 1) or benign (BI-RADS 2) microcalcifications. The remaining exhibited suspicious findings (BI-RADS 4-5) in the recent image. Mammograms cannot be directly subtracted, due to tissue changes over time and breast deformation during mammography. To overcome this challenge, optimised preprocessing, image registration, and postprocessing procedures were developed. Machine learning techniques were employed to eliminate false positives (normal tissue misclassified as microcalcifications) and to classify the true microcalcifications as BI-RADS benign or suspicious. Ninety-six features were extracted and nine classifiers were evaluated with and without temporal subtraction. The performance was assessed by measuring sensitivity, specificity, accuracy, and area under the curve (AUC) at receiver operator characteristics analysis. RESULTS: Using temporal subtraction, the contrast ratio improved ~ 57 times compared to the most recent mammograms, enhancing the detection of the radiologic changes. Classifying as BI-RADS benign versus suspicious microcalcifications, resulted in 90.3% accuracy and 0.87 AUC, compared to 82.7% and 0.81 using just the most recent mammogram (p = 0.003). CONCLUSION: Compared to using the most recent mammogram alone, temporal subtraction is more effective in the microcalcifications detection and classification and may play a role in automated diagnosis systems.

Vesseg: An Open-Source Tool for Deep Learning-Based Atherosclerotic Plaque Quantification in Histopathology Images-Brief Report.

Objective: Manual plaque segmentation in microscopy images is a time-consuming process in atherosclerosis research and potentially subject to unacceptable user-to-user variability and observer bias. We address this by releasing Vesseg a tool that includes state-of-the-art deep learning models for atherosclerotic plaque segmentation. Approach and Results: Vesseg is a containerized, extensible, open-source, and user-oriented tool. It includes 2 models, trained and tested on 1089 hematoxylin-eosin stained mouse model atherosclerotic brachiocephalic artery sections. The models were compared to 3 human raters. Vesseg can be accessed at https://vesseg .online or downloaded. The models show mean Soerensen-Dice scores of 0.91+/-0.15 for plaque and 0.97+/-0.08 for lumen pixels. The mean accuracy is 0.98+/-0.05. Vesseg is already in active use, generating time savings of >10 minutes per slide. Conclusions: Vesseg brings state-of-the-art deep learning methods to atherosclerosis research, providing drastic time savings, while allowing for continuous improvement of models and the underlying pipeline.

Texture analysis based on quantitative magnetic resonance imaging to assess kidney function: a preliminary study.

BACKGROUND: Magnetic resonance imaging (MRI) has demonstrated its potential in the evaluation of renal function. Texture analysis (TA) is a novel technique to quantify tissue heterogeneity. We aim to investigate the feasibility of using TA based on the apparent diffusion coefficient (ADC), as well as T1 and T2 maps to evaluate renal function. METHODS: Patients with impaired renal function and subjects with a normal renal function who underwent renal diffusion weighted imaging (DWI), as well as T1 and T2 mapping at 3T, were prospectively enrolled. The participants were classified into four groups according to the estimated glomerular filtration rate (eGFR, mL/min/1.73 m2): normal (eGFR ≥90), mildly impaired (60≤ eGFR <90), moderately impaired (30≤ eGFR <60), and severely impaired (eGFR <30) renal function groups. Texture features quantified from the renal cortex or medulla were selected to build classifiers to discriminate different renal function groups by plotting receiver operating characteristic (ROC) curves and calculating the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). RESULTS: In total, 116 candidates were included (94 patients and 22 healthy volunteers, mean age 37.9±14.9 years). There were 46 participants in the normal renal function group, 14 in the mildly impaired renal function group, 27 in the moderately impaired renal function group, and 29 in the severely impaired renal function group. Texture features from the ADC and T1 maps exhibited a good correlation to eGFR. The AUC, sensitivity, specificity, PPV, and NPV to differentiate between the normal and impaired renal function groups were 0.835, 0.792, 0.867, 0.905, and 0.722, respectively; to differentiate between the mildly impaired and moderately impaired groups were 0.937, 0.889, 0.857, 0.923, and 0.800, respectively; and to differentiate between the moderately impaired and severely impaired groups was 0.940, 0.759, 0.889, 0.880, and 0.774, respectively. CONCLUSIONS: TA based on ADC and T1 maps is feasible for evaluating renal function with relatively good accuracy.

Radiomic analysis of lung cancer for the assessment of patient prognosis and intratumor heterogeneity.

OBJECTIVE: To determine whether the radiomic features of lung lesions on computed tomography correlate with overall survival in lung cancer patients. MATERIALS AND METHODS: This was a retrospective study involving 101 consecutive patients with malignant neoplasms confirmed by biopsy or surgery. On computed tomography images, the lesions were submitted to semi-automated segmentation and were characterized on the basis of 2,465 radiomic variables. The prognostic assessment was based on Kaplan-Meier analysis and log-rank tests, according to the median value of the radiomic variables. RESULTS: Of the 101 patients evaluated, 28 died (16 dying from lung cancer), and 73 were censored, with a mean overall survival time of 1,819.4 days (95% confidence interval [95% CI]: 1,481.2-2,157.5). One radiomic feature (the mean of the Fourier transform) presented a difference on Kaplan-Meier curves (p < 0.05). A high-risk group of patients was identified on the basis of high values for the mean of the Fourier transform. In that group, the mean survival time was 1,465.4 days (95% CI: 985.2-1,945.6), with a hazard ratio of 2.12 (95% CI: 1.01-4.48). We also identified a low-risk group, in which the mean of the Fourier transform was low (mean survival time of 2,164.8 days; 95% CI: 1,745.4-2,584.1). CONCLUSION: A radiomic signature based on the Fourier transform correlates with overall survival, representing a prognostic biomarker for risk stratification in patients with lung cancer.

OBJETIVO: Associar características radiômicas de lesões pulmonares em imagens de tomografia computadorizada com a sobrevida global de pacientes com câncer de pulmão. MATERIAIS E MÉTODOS: Estudo retrospectivo composto por 101 pacientes consecutivos com neoplasia maligna confirmada por biópsia/cirurgia. As lesões foram semiautomaticamente segmentadas e caracterizadas por 2.465 variáveis radiômicas. A avaliação prognóstica foi baseada na análise de Kaplan-Meier e no teste log-rank, de acordo com a mediana dos valores das variáveis. RESULTADOS: Vinte e oito pacientes faleceram (16 por câncer de pulmão) e 73 foram censurados, com tempo médio de sobrevida de 1.819,4 dias (intervalo de confiança 95% [IC 95%]: 1.481,2-2.157,5). Uma característica radiômica (média de Fourier) apresentou diferença nas curvas de Kaplan-Meier (p < 0,05). Um grupo de pacientes de maior risco foi identificado a partir de valores altos da variável: sobrevida de 1.465,4 dias (IC 95%: 985,2-1.945,6) e razão de risco de 2,12 (IC 95%: 1,01-4,48). Um grupo de menor risco foi identificado a partir de valores baixos da variável (sobrevida de 2.164,8 dias; IC 95%: 1.745,4-2.584,1). CONCLUSÃO: Este estudo apresentou uma assinatura radiômica em imagens de tomografia computadorizada, baseada na transformada de Fourier, correlacionada com a sobrevida global de pacientes com câncer de pulmão, representando assim um biomarcador prognóstico.

Radiomic analysis of lung cancer for the assessment of patient prognosis and intratumor heterogeneity / Análise radiômica do câncer de pulmão para avaliação prognóstica do paciente e da heterogeneidade intratumoral

Abstract Objective: To determine whether the radiomic features of lung lesions on computed tomography correlate with overall survival in lung cancer patients. Materials and Methods: This was a retrospective study involving 101 consecutive patients with malignant neoplasms confirmed by biopsy or surgery. On computed tomography images, the lesions were submitted to semi-automated segmentation and were characterized on the basis of 2,465 radiomic variables. The prognostic assessment was based on Kaplan-Meier analysis and log-rank tests, according to the median value of the radiomic variables. Results: Of the 101 patients evaluated, 28 died (16 dying from lung cancer), and 73 were censored, with a mean overall survival time of 1,819.4 days (95% confidence interval [95% CI]: 1,481.2-2,157.5). One radiomic feature (the mean of the Fourier transform) presented a difference on Kaplan-Meier curves (p < 0.05). A high-risk group of patients was identified on the basis of high values for the mean of the Fourier transform. In that group, the mean survival time was 1,465.4 days (95% CI: 985.2-1,945.6), with a hazard ratio of 2.12 (95% CI: 1.01-4.48). We also identified a low-risk group, in which the mean of the Fourier transform was low (mean survival time of 2,164.8 days; 95% CI: 1,745.4-2,584.1). Conclusion: A radiomic signature based on the Fourier transform correlates with overall survival, representing a prognostic biomarker for risk stratification in patients with lung cancer.

Resumo Objetivo: Associar características radiômicas de lesões pulmonares em imagens de tomografia computadorizada com a sobrevida global de pacientes com câncer de pulmão. Materiais e Métodos: Estudo retrospectivo composto por 101 pacientes consecutivos com neoplasia maligna confirmada por biópsia/cirurgia. As lesões foram semiautomaticamente segmentadas e caracterizadas por 2.465 variáveis radiômicas. A avaliação prognóstica foi baseada na análise de Kaplan-Meier e no teste log-rank, de acordo com a mediana dos valores das variáveis. Resultados: Vinte e oito pacientes faleceram (16 por câncer de pulmão) e 73 foram censurados, com tempo médio de sobrevida de 1.819,4 dias (intervalo de confiança 95% [IC 95%]: 1.481,2-2.157,5). Uma característica radiômica (média de Fourier) apresentou diferença nas curvas de Kaplan-Meier (p < 0,05). Um grupo de pacientes de maior risco foi identificado a partir de valores altos da variável: sobrevida de 1.465,4 dias (IC 95%: 985,2-1.945,6) e razão de risco de 2,12 (IC 95%: 1,01-4,48). Um grupo de menor risco foi identificado a partir de valores baixos da variável (sobrevida de 2.164,8 dias; IC 95%: 1.745,4-2.584,1). Conclusão: Este estudo apresentou uma assinatura radiômica em imagens de tomografia computadorizada, baseada na transformada de Fourier, correlacionada com a sobrevida global de pacientes com câncer de pulmão, representando assim um biomarcador prognóstico.