A convolutional deep learning model for improving mammographic breast-microcalcification diagnosis

AbstractThis study aimed to assess the diagnostic performance of deep convolutional neural networks (DCNNs) in classifying breast microcalcification in screening mammograms. To this end, 1579 mammographic images were collected retrospectively from patients exhibiting suspicious microcalcification in screening mammograms between July 2007 and December 2019. Five pre-trained DCNN models and an ensemble model were used to classify the microcalcifications as either malignant or benign. Approximately one million images from the ImageNet database had been used to train the five DCNN models. Herein, 1121 mammographic images were used for individual model fine-tuning, 198 for validation, and 260 for testing. Gradient-weighted class activation mapping (Grad-CAM) was used to confirm the validity of the DCNN models in highlighting the microcalcification regions most critical for determining the final class. The ensemble model yielded the best AUC (0.856). The DenseNet-201 model achieved the best sensitivity (82.47%) and negative predictive value (NPV; 86.92%). The ResNet-101 model yielded the best accuracy (81.54%), specificity (91.41%), and positive predictive value (PPV; 81.82%). The high PPV and specificity achieved by the ResNet-101 model, in particular, demonstrated the model effectiveness in microcalcification diagnosis, which, in turn, may considerably help reduce unnecessary biopsies.

Download Full-text

Race Recognition Using Deep Convolutional Neural Networks

Symmetry ◽

10.3390/sym10110564 ◽

2018 ◽

Vol 10 (11) ◽

pp. 564 ◽

Cited By ~ 11

Author(s):

Thanh Vo ◽

Trang Nguyen ◽

C. Le

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Object Recognition ◽

Difficult Problem ◽

Fine Tuning ◽

Surveillance Systems ◽

Deep Convolutional Neural Networks ◽

Facebook Pages ◽

Deep Learning Model

Race recognition (RR), which has many applications such as in surveillance systems, image/video understanding, analysis, etc., is a difficult problem to solve completely. To contribute towards solving that problem, this article investigates using a deep learning model. An efficient Race Recognition Framework (RRF) is proposed that includes information collector (IC), face detection and preprocessing (FD&P), and RR modules. For the RR module, this study proposes two independent models. The first model is RR using a deep convolutional neural network (CNN) (the RR-CNN model). The second model (the RR-VGG model) is a fine-tuning model for RR based on VGG, the famous trained model for object recognition. In order to examine the performance of our proposed framework, we perform an experiment on our dataset named VNFaces, composed specifically of images collected from Facebook pages of Vietnamese people, to compare the accuracy between RR-CNN and RR-VGG. The experimental results show that for the VNFaces dataset, the RR-VGG model with augmented input images yields the best accuracy at 88.87% while RR-CNN, an independent and lightweight model, yields 88.64% accuracy. The extension experiments conducted prove that our proposed models could be applied to other race dataset problems such as Japanese, Chinese, or Brazilian with over 90% accuracy; the fine-tuning RR-VGG model achieved the best accuracy and is recommended for most scenarios.

Download Full-text

Predictive Evaluation on Cytological Sample of Metastatic Melanoma: The Role of BRAF Immunocytochemistry in the Molecular Era

Diagnostics ◽

10.3390/diagnostics11061110 ◽

2021 ◽

Vol 11 (6) ◽

pp. 1110

Author(s):

Andrea Ronchi ◽

Marco Montella ◽

Federica Zito Marino ◽

Michele Caraglia ◽

Anna Grimaldi ◽

...

Keyword(s):

Metastatic Melanoma ◽

Molecular Analysis ◽

Predictive Value ◽

Diagnostic Performance ◽

Cost Effective ◽

Needle Aspiration ◽

Braf V600e ◽

Mutational Status ◽

Needle Aspiration Cytology ◽

Sensitivity Specificity

Background: Cutaneous malignant melanoma is an aggressive neoplasm. In advanced cases, the therapeutic choice depends on the mutational status of BRAF. Fine needle aspiration cytology (FNA) is often applied to the management of patients affected by melanoma, mainly for the diagnosis of metastases. The evaluation of BRAF mutational status by sequencing technique on cytological samples may be inconvenient, as it is a time and biomaterial-consuming technique. Recently, BRAF immunocytochemistry (ICC) was applied for the evaluation of BRAF V600E mutational status. Although it may be useful mainly in cytological samples, data about BRAF ICC on cytological samples are missing. Methods: We performed BRAF ICC on a series of 50 FNA samples of metastatic melanoma. BRAF molecular analysis was performed on the same cytological samples or on the corresponding histological samples. Molecular analysis was considered the gold standard. Results: BRAF ICC results were adequate in 49 out of 50 (98%) cases, positive in 15 out of 50 (30%) cases and negative in 34 out of 50 (68%) of cases. Overall, BRAF ICC sensitivity, specificity, positive predictive value and negative predictive value results were 88.2%, 100%, 100% and 94.1%, respectively. The diagnostic performance of BRAF ICC results was perfect when molecular evaluation was performed on the same cytological samples. Hyperpigmentation represents the main limitation of the technique. Conclusions: BRAF ICC is a rapid, cost-effective method for detecting BRAF V600E mutation in melanoma metastases, applicable with high diagnostic performance to cytological samples. It could represent the first step to evaluate BRAF mutational status in cytological samples, mainly in poorly cellular cases.

Download Full-text

A Rapid Amniotic Fluid Interleukin-6 Assessment for the Identification of Intra-Amniotic Inflammation in Women with Preterm Labor and Intact Membranes

Fetal Diagnosis and Therapy ◽

10.1159/000513660 ◽

2021 ◽

pp. 1-6

Author(s):

Teresa Cobo ◽

Victoria Aldecoa ◽

Magdalena Holeckova ◽

Ctirad Andrys ◽

Xavier Filella ◽

...

Keyword(s):

Amniotic Fluid ◽

Predictive Model ◽

Predictive Value ◽

Preterm Labor ◽

Diagnostic Performance ◽

Clinical Signs ◽

Entire Study ◽

Microbial Invasion ◽

Automated Method ◽

Spontaneous Delivery

Objectives: A multivariable predictive model has recently been developed with good accuracy to predict spontaneous preterm delivery within 7 days in women with preterm labor (PTL) and intact membranes. However, this model measures amniotic fluid (AF) interleukin (IL)-6 concentrations using the ELISA method, thereby limiting clinical implementation. The main objectives of this study were to validate the automated immunoassay as a quantitative method to measure AF IL-6 in women with PTL and to evaluate the diagnostic performance of AF IL-6 alone and as part of a multivariable predictive model to predict spontaneous delivery in 7 days with this automated method. Study Design: This is a retrospective observational study in women with PTL below 34 weeks who underwent amniocentesis to rule out microbial invasion of the amniotic cavity. Women with clinical signs of chorioamnionitis, cervical length measurement at admission >5th centile, maternal age <18 years, and no consent to perform amniocentesis for this indication were excluded. The local Institutional Review Boards approved the study (HCB/2019/0940). Analysis of AF IL-6 Concentrations: AF IL-6 concentrations were measured using an automated Cobas e602 electrochemiluminescence immunoanalyzer and Human IL-6 Quantikine ELISA kit. Results: Of the entire study group (n = 100), 38 women spontaneously delivered within 7 days after admission. Both laboratory methods showed good agreement (intraclass correlation coefficient: 0.937 (95% confidence interval [CI] 0.908–0.957); p < 0.001). Diagnostic performance of AF IL-6 to predict spontaneous delivery within 7 days when it was included in the multivariable predictive model showed an area under the receiver operating characteristic curve of 0.894 (95% CI 0.799–0.955), sensitivity of 97%, specificity of 74%, positive predictive value of 73%, negative predictive value of 97%, positive likelihood ratio (LR) of 3.7, and negative LR of 0.045. Conclusion: While both analytical methods were comparable for measuring AF IL-6 concentrations in women with PTL, the Cobas immunoanalyzer provided rapid diagnosis of intra-amniotic inflammation within minutes. The predictive model showed a good diagnostic performance to target women at high risk of spontaneous delivery within 7 days.

Download Full-text

Interpretable deep learning for the remote characterisation of ambulation in multiple sclerosis using smartphones

Scientific Reports ◽

10.1038/s41598-021-92776-x ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Andrew P. Creagh ◽

Florian Lipsmeier ◽

Michael Lindemann ◽

Maarten De Vos

Keyword(s):

Multiple Sclerosis ◽

Deep Learning ◽

Inertial Sensor ◽

Heterogeneous Data ◽

Fine Tuning ◽

Sensor Data ◽

Support Vector ◽

Deep Convolutional Neural Networks ◽

Healthcare Applications ◽

Feature Based

AbstractThe emergence of digital technologies such as smartphones in healthcare applications have demonstrated the possibility of developing rich, continuous, and objective measures of multiple sclerosis (MS) disability that can be administered remotely and out-of-clinic. Deep Convolutional Neural Networks (DCNN) may capture a richer representation of healthy and MS-related ambulatory characteristics from the raw smartphone-based inertial sensor data than standard feature-based methodologies. To overcome the typical limitations associated with remotely generated health data, such as low subject numbers, sparsity, and heterogeneous data, a transfer learning (TL) model from similar large open-source datasets was proposed. Our TL framework leveraged the ambulatory information learned on human activity recognition (HAR) tasks collected from wearable smartphone sensor data. It was demonstrated that fine-tuning TL DCNN HAR models towards MS disease recognition tasks outperformed previous Support Vector Machine (SVM) feature-based methods, as well as DCNN models trained end-to-end, by upwards of 8–15%. A lack of transparency of “black-box” deep networks remains one of the largest stumbling blocks to the wider acceptance of deep learning for clinical applications. Ensuing work therefore aimed to visualise DCNN decisions attributed by relevance heatmaps using Layer-Wise Relevance Propagation (LRP). Through the LRP framework, the patterns captured from smartphone-based inertial sensor data that were reflective of those who are healthy versus people with MS (PwMS) could begin to be established and understood. Interpretations suggested that cadence-based measures, gait speed, and ambulation-related signal perturbations were distinct characteristics that distinguished MS disability from healthy participants. Robust and interpretable outcomes, generated from high-frequency out-of-clinic assessments, could greatly augment the current in-clinic assessment picture for PwMS, to inform better disease management techniques, and enable the development of better therapeutic interventions.

Download Full-text

Comparison of the Diagnostic Performance of Novel Slim Biopsy Forceps with Conventional Biopsy Forceps for Biliary Stricture: A Multicenter Retrospective Study

Journal of Personalized Medicine ◽

10.3390/jpm11010055 ◽

2021 ◽

Vol 11 (1) ◽

pp. 55

Author(s):

Eun Suk Jung ◽

Se Woo Park ◽

Jung Hee Kim ◽

Jang Han Jung ◽

Min Jae Yang ◽

...

Keyword(s):

Adverse Events ◽

Predictive Value ◽

Diagnostic Performance ◽

Biliary Stricture ◽

Accurate Diagnosis ◽

The Novel ◽

Biopsy Forceps ◽

Tertiary Hospitals ◽

Tissue Acquisition ◽

Histological Confirmation

Novel slim biopsy forceps provide some technical advantages to facilitate a more accurate diagnosis, although we are not aware of any comparative studies. Therefore, we compared tissue acquisition and diagnostic accuracy between novel slim biopsy forceps and conventional biopsy forceps in cases with a biliary stricture. We reviewed 341 patients who underwent endoscopic retrograde cholangiopancreatography for the histological confirmation of biliary stricture at two tertiary hospitals between 2013 and 2020. The primary endpoint was the forceps’ diagnostic accuracies. We included 276 patients who underwent biopsy using the novel forceps (n = 130) or conventional forceps (n = 146). The novel forceps provided 81.7% sensitivity, 100.0% specificity, positive-predictive value (PPV) of 100.0%, and negative-predictive value (NPV) of 57.8%, with an accuracy of 85.4% when the diagnosis by endobiliary biopsy included suspected or positive malignancy. The conventional forceps provided 61.7% sensitivity, 100.0% specificity, PPV of 100.0%, and NPV of 36.1%, with an accuracy of 68.5%. Only novel forceps use was significantly associated with an accurate diagnosis (odds ratio: 2.70, 95% confidence interval: 1.52–5.00). There were no significant inter-group differences in the procedure-related rates of adverse events. Endobiliary biopsy using novel forceps offered better diagnostic performance and more acceptable procedure-related adverse events than conventional forceps.

Download Full-text

P6175Prediction of coronary revascularization by coronary computed tomography angiography derived fractional flow reserve - different algorithms for interpretation

European Heart Journal ◽

10.1093/eurheartj/ehz746.0781 ◽

2019 ◽

Vol 40 (Supplement_1) ◽

Author(s):

K T Madsen ◽

K T Veien ◽

B L Noergaard ◽

P Larsen ◽

L Deibjerg ◽

...

Keyword(s):

Chest Pain ◽

Fractional Flow Reserve ◽

Predictive Value ◽

Diagnostic Performance ◽

Coronary Revascularization ◽

Standard Of Care ◽

Fractional Flow ◽

Coronary Cta ◽

Flow Reserve ◽

Stable Chest Pain

Abstract Introduction Coronary CT angiography (CTA) derived fractional flow reserve (FFRct) is increasingly used for guiding referral to invasive procedures in patients with stable chest pain. However, optimal interpretation of FFRct-analysis in terms of location and threshold of applied FFRct-values is unclear. Purpose To evaluate the clinical performance of various vessel-specific physiological FFRct derived measures of ischemia for prediction of standard of care guided coronary revascularization in patients with stable chest pain and coronary artery disease as determined by coronary CTA. Methods Retrospective study in patients with stable chest pain referred for coronary angiography based on coronary CTA. Standard acquired coronary CTA data sets were transmitted for core-laboratory analysis at HeartFlow. Any FFRct value in the major coronary arteries ≥1.8 mm in diameter, including side branches, were registered. Lesions were categorized as positive for ischemia using 6 different algorithms: Lowest in vessel FFRct-value (1) ≤0.75 or (2) ≤0.80; 2 cm distal-to-lesion FFRct-value (3) ≤0.75 or (4) ≤0.80; ΔFFRct (5) ≥0.06 or a combination of 2 and 5. The personnel responsible for downstream patient management had no information regarding FFRct test results. Results A total of 172 patients were included. Revascularization was performed in 62 (35%) patients. The diagnostic performance of different FFRct algorithms for predicting standard of care guided coronary revascularization is shown in the Table. Revascularization Predictions by FFRct N=172 Diagnostic performance FFRCT false negative FFRCT false positive Values given as (%) No. of revasc vessels No. of abnormal vessels FFRCT Algorithm Sens Spec PPV NPV Acc 1 2 3 1 2 3 Distal FFRCT ≤0.75 77 68 58 84 72 12 2 0 29 5 1 Distal FFRCT ≤0.80 92 43 48 90 61 5 0 0 40 20 3 Lesion-specific FFRCT ≤0.75 68 86 74 83 80 17 3 0 12 3 0 Lesion-specific FFRCT ≤0.80 82 78 68 89 80 10 2 0 21 3 1 ΔFFRCT ≥0.06 98 36 47 98 59 1 0 0 51 19 0 Combinationa 92 54 53 92 67 5 0 0 39 12 0 aDistal FFRCT ≤0.80 and ΔFFRCT ≥0.06. Sens = sensitivity; Spec = specificity; PPV = positive predictive value; NPV = negative predictive value; Acc = accuracy; FFRCT = fractional flow reserve derived from coronary CTA; ΔFFRCT = difference between FFRCT-value immediately proximal and distal to lesion; Revasc = revascularized. Conclusion The diagnostic performance of FFRct in terms of predicting standard of care guided coronary revascularization is dependent on the applied algorithm for interpretation of the FFRct-analysis.

Download Full-text

Aerial Imagery Paddy Seedlings Inspection Using Deep Learning

Remote Sensing ◽

10.3390/rs14020274 ◽

2022 ◽

Vol 14 (2) ◽

pp. 274

Author(s):

Mohamed Marzhar Anuar ◽

Alfian Abdul Halin ◽

Thinagaran Perumal ◽

Bahareh Kalantar

Keyword(s):

Computer Vision ◽

Deep Learning ◽

Production Costs ◽

Fine Tuning ◽

Aerial Imagery ◽

Planting Density ◽

Deep Convolutional Neural Networks ◽

Security Issues ◽

Agriculture Industry ◽

Paddy Cultivation

In recent years complex food security issues caused by climatic changes, limitations in human labour, and increasing production costs require a strategic approach in addressing problems. The emergence of artificial intelligence due to the capability of recent advances in computing architectures could become a new alternative to existing solutions. Deep learning algorithms in computer vision for image classification and object detection can facilitate the agriculture industry, especially in paddy cultivation, to alleviate human efforts in laborious, burdensome, and repetitive tasks. Optimal planting density is a crucial factor for paddy cultivation as it will influence the quality and quantity of production. There have been several studies involving planting density using computer vision and remote sensing approaches. While most of the studies have shown promising results, they have disadvantages and show room for improvement. One of the disadvantages is that the studies aim to detect and count all the paddy seedlings to determine planting density. The defective paddy seedlings’ locations are not pointed out to help farmers during the sowing process. In this work we aimed to explore several deep convolutional neural networks (DCNN) models to determine which one performs the best for defective paddy seedling detection using aerial imagery. Thus, we evaluated the accuracy, robustness, and inference latency of one- and two-stage pretrained object detectors combined with state-of-the-art feature extractors such as EfficientNet, ResNet50, and MobilenetV2 as a backbone. We also investigated the effect of transfer learning with fine-tuning on the performance of the aforementioned pretrained models. Experimental results showed that our proposed methods were capable of detecting the defective paddy rice seedlings with the highest precision and an F1-Score of 0.83 and 0.77, respectively, using a one-stage pretrained object detector called EfficientDet-D1 EficientNet.

Download Full-text

A Novel Model Based on Deep Convolutional Neural Network Improves Diagnostic Accuracy of Intramucosal Gastric Cancer (With Video)

Frontiers in Oncology ◽

10.3389/fonc.2021.622827 ◽

2021 ◽

Vol 11 ◽

Author(s):

Dehua Tang ◽

Jie Zhou ◽

Lei Wang ◽

Muhan Ni ◽

Min Chen ◽

...

Keyword(s):

Gastric Cancer ◽

Diagnostic Accuracy ◽

Diagnostic Performance ◽

Training Dataset ◽

Deep Convolutional Neural Networks ◽

Testing Dataset ◽

Kappa Value ◽

The Mean ◽

Gastric Cancer Patients ◽

Novel Model

Background and AimsPrediction of intramucosal gastric cancer (GC) is a big challenge. It is not clear whether artificial intelligence could assist endoscopists in the diagnosis.MethodsA deep convolutional neural networks (DCNN) model was developed via retrospectively collected 3407 endoscopic images from 666 gastric cancer patients from two Endoscopy Centers (training dataset). The DCNN model’s performance was tested with 228 images from 62 independent patients (testing dataset). The endoscopists evaluated the image and video testing dataset with or without the DCNN model’s assistance, respectively. Endoscopists’ diagnostic performance was compared with or without the DCNN model’s assistance and investigated the effects of assistance using correlations and linear regression analyses.ResultsThe DCNN model discriminated intramucosal GC from advanced GC with an AUC of 0.942 (95% CI, 0.915–0.970), a sensitivity of 90.5% (95% CI, 84.1%–95.4%), and a specificity of 85.3% (95% CI, 77.1%–90.9%) in the testing dataset. The diagnostic performance of novice endoscopists was comparable to those of expert endoscopists with the DCNN model’s assistance (accuracy: 84.6% vs. 85.5%, sensitivity: 85.7% vs. 87.4%, specificity: 83.3% vs. 83.0%). The mean pairwise kappa value of endoscopists was increased significantly with the DCNN model’s assistance (0.430–0.629 vs. 0.660–0.861). The diagnostic duration reduced considerably with the assistance of the DCNN model from 4.35s to 3.01s. The correlation between the perseverance of effort and diagnostic accuracy of endoscopists was diminished using the DCNN model (r: 0.470 vs. 0.076).ConclusionsAn AI-assisted system was established and found useful for novice endoscopists to achieve comparable diagnostic performance with experts.

Download Full-text

Diagnostic Performance and Reliability of the Standardized Computed Tomography Reporting System for Acute Appendicitis: Experience in a Tertiary Care Academic Center

Journal of the Medical Association of Thailand ◽

10.35755/jmedassocthai.2021.07.12434 ◽

2021 ◽

Vol 104 (7) ◽

pp. 1102-1108

Keyword(s):

Computed Tomography ◽

Acute Appendicitis ◽

Diagnostic Accuracy ◽

Predictive Value ◽

Diagnostic Performance ◽

Interobserver Agreement ◽

Reporting System ◽

Lower Quadrant ◽

Suspected Appendicitis ◽

Diagnostic Certainty

Background: Computed tomography (CT) is generally accepted as a modality of choice for imaging workup in patients with suspected appendicitis. A standardized CT reporting system, CT certainty score, has been proposed to improve diagnostic accuracy and to reduce ambiguous CT reports. Objective: To assess the diagnostic performance and the reliability of the standardized CT reporting system for acute appendicitis in Thai adults. Materials and Methods: The present study was a retrospective data review of 421 adult patients who had CT scans of the appendix between January 2016 and December 2017. The clinical and imaging data were extracted and analyzed. The pathological result was used as a standard of reference. The diagnostic performance and interobserver agreement of the standardized CT reporting system were estimated. Results: One hundred sixty-three patients, with a mean age of 41.7 years, had clinical diagnoses of acute appendicitis. Using standardized CT report, radiologists were highly accurate at diagnosing appendicitis [area under curve (AUC) 0.988 (95% CI 0.98 to 1.00); p<0.001]. The estimated sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 95.1% (95% CI 90.6 to 97.9), 95.7% (95% CI 92.5 to 97.9), 93.4% (95% CI 88.7 to 96.2), 96.9% (95% CI 93.0 to 97.2), 95.5% (95% CI 93.0 to 97.3), respectively. The interobserver agreement was greater than 80% for all binary objective findings and more than 90% agreement on the presence or absence of greater-than-3-mm wall thickness, appendicolith, periappendiceal air, and right lower quadrant fluid collection. The use of CT certainty score had interobserver agreement of 78% (κ=0.69; 95% CI 0.62 to 0.77). Conclusion: Using a standardized CT reporting system yielded a high diagnostic accuracy and high reproducibility of supportive CT findings for appendicitis in at-risk patients. The standardized CT reporting system can improve diagnostic certainty, accuracy, and guide patient management. Keywords: Appendicitis; Certainty score; Computed tomography; Standardized reporting system

Download Full-text

Abstract P319: Can Deep Learning Find the Ischemic Core on CT? Transfer Learning From Pre-Trained MRI-Based Networks

Stroke ◽

10.1161/str.52.suppl_1.p319 ◽

2021 ◽

Vol 52 (Suppl_1) ◽

Author(s):

Yannan Yu ◽

Soren Christensen ◽

Yuan Xie ◽

Enhao Gong ◽

Maarten G Lansberg ◽

...

Keyword(s):

Deep Learning ◽

Ground Truth ◽

Learning Model ◽

Fine Tuning ◽

Learning Models ◽

Starting Point ◽

Stroke Lesion ◽

Ischemic Core ◽

Deep Learning Model

Objective: Ischemic core prediction from CT perfusion (CTP) remains inaccurate compared with gold standard diffusion-weighted imaging (DWI). We evaluated if a deep learning model to predict the DWI lesion from MR perfusion (MRP) could facilitate ischemic core prediction on CTP. Method: Using the multi-center CRISP cohort of acute ischemic stroke patient with CTP before thrombectomy, we included patients with major reperfusion (TICI score≥2b), adequate image quality, and follow-up MRI at 3-7 days. Perfusion parameters including Tmax, mean transient time, cerebral blood flow (CBF), and cerebral blood volume were reconstructed by RAPID software. Core lab experts outlined the stroke lesion on the follow-up MRI. A previously trained MRI model in a separate group of patients was used as a starting point, which used MRP parameters as input and RAPID ischemic core on DWI as ground truth. We fine-tuned this model, using CTP parameters as input, and follow-up MRI as ground truth. Another model was also trained from scratch with only CTP data. 5-fold cross validation was used. Performance of the models was compared with ischemic core (rCBF≤30%) from RAPID software to identify the presence of a large infarct (volume>70 or >100ml). Results: 94 patients in the CRISP trial met the inclusion criteria (mean age 67±15 years, 52% male, median baseline NIHSS 18, median 90-day mRS 2). Without fine-tuning, the MRI model had an agreement of 73% in infarct >70ml, and 69% in >100ml; the MRI model fine-tuned on CT improved the agreement to 77% and 73%; The CT model trained from scratch had agreements of 73% and 71%; All of the deep learning models outperformed the rCBF segmentation from RAPID, which had agreements of 51% and 64%. See Table and figure. Conclusions: It is feasible to apply MRP-based deep learning model to CT. Fine-tuning with CTP data further improves the predictions. All deep learning models predict the stroke lesion after major recanalization better than thresholding approaches based on rCBF.

Download Full-text