Deep learning based automatic segmentation of the Internal Pudendal Artery in definitive radiotherapy treatment planning of localized prostate cancer.

Background and purpose: Radiation-induced erectile dysfunction (RiED) commonly affects prostate cancer patients, prompting clinical trials across institutions to explore dose-sparing to internal-pudendal-arteries (IPA) for preserving sexual potency. IPA, challenging to segment, isn't conventionally considered an organ-at-risk (OAR). This study proposes a deep learning (DL) auto-segmentation model for IPA, using Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) or CT alone to accommodate varied clinical practices. Materials and methods: A total of 86 patients with CT and MRI images and noisy IPA labels were recruited in this study. We split the data into 42/14/30 for model training, testing, and a clinical observer study, respectively. There were three major innovations in this model: 1) we designed an architecture with squeeze-and-excite blocks and modality attention for effective feature extraction and production of accurate segmentation, 2) a novel loss function was used for training the model effectively with noisy labels, and 3) modality dropout strategy was used for making the model capable of segmentation in the absence of MRI. Results: Test dataset metrics were DSC 61.71 ± 7.7 %, ASD 2.5 ± .87 mm, and HD95 7.0 ± 2.3 mm. AI segmented contours showed dosimetric similarity to expert physician's contours. Observer study indicated higher scores for AI contours (mean = 3.7) compared to inexperienced physicians' contours (mean = 3.1). Inexperienced physicians improved scores to 3.7 when starting with AI contours. Conclusion: The proposed model achieved good quality IPA contours to improve uniformity of segmentation and to facilitate introduction of standardized IPA segmentation into clinical trials and practice.

In silico evaluation and feasibility of near margin-less head and neck daily adaptive radiotherapy.

OBJECTIVE: We explore the potential dosimetric benefits of reducing treatment volumes through daily adaptive radiation therapy for head and neck cancer (HNC) patients using the Ethos system/Intelligent Optimizer Engine (IOE). We hypothesize reducing treatment volumes afforded by daily adaption will significantly reduce the dose to adjacent organs at risk. We also explore the capability of the Ethos IOE to accommodate this highly conformal approach in HNC radiation therapy. METHODS: Ten HNC patients from a phase II trial were chosen, and their cone-beam CT (CBCT) scans were uploaded to the adaptive RT (ART) emulator. A new initial reference plan was generated using both a 1 mm and 5 mm planning target volume (PTV) expansion. Daily adaptive ART plans (1 mm) were simulated from the clinical CBCT taken every fifth fraction. Additionally, using physician-modified ART contours the larger 5 mm plan was recalculated on this recontoured on daily anatomy. Changes in target and OAR contours were measured using Dice coefficients as a surrogate of clinician effort. PTV coverage and organ-at-risk (OAR) doses were statistically compared, and the robustness of each ART plan was evaluated at fractions 5 and 35 to observe if OAR doses were within 3 Gy of pre-plan. RESULTS: This study involved six patients with oropharynx and four with larynx cancer, totaling 70 adaptive fractions. The primary and nodal gross tumor volumes (GTV) required the most adjustments, with median Dice scores of 0.88 (range: 0.80-0.93) and 0.83 (range: 0.66-0.91), respectively. For the 5th and 35th fraction plans, 80 % of structures met robustness criteria (quartile 1-3: 67-100 % and 70-90 %). Adaptive planning improved median PTV V100% coverage for doses of 70 Gy (96 % vs. 95.6 %), 66.5 Gy (98.5 % vs. 76.5 %), and 63 Gy (98.9 % vs. 74.9 %) (p < 0.03). Implementing ART with total volume reduction yielded median dose reductions of 7-12 Gy to key organs-at-risk (OARs) like submandibular glands, parotids, oral cavity, and constrictors (p < 0.05). CONCLUSIONS: The IOE enables feasible daily ART treatments with reduced margins while enhancing target coverage and reducing OAR doses for HNC patients. A phase II trial recently finished accrual and forthcoming analysis will determine if these dosimetric improvements correlate with improved patient-reported outcomes.

The accuracy of artificial intelligence deformed nodal structures in cervical online cone-beam-based adaptive radiotherapy.

Background and Purpose: Online cone-beam-based adaptive radiotherapy (ART) adjusts for anatomical changes during external beam radiotherapy. However, limited cone-beam image quality complicates nodal contouring. Despite this challenge, artificial-intelligence guided deformation (AID) can auto-generate nodal contours. Our study investigated the optimal use of such contours in cervical online cone-beam-based ART. Materials and Methods: From 136 adaptive fractions across 21 cervical cancer patients with nodal disease, we extracted 649 clinically-delivered and AID clinical target volume (CTV) lymph node boost structures. We assessed geometric alignment between AID and clinical CTVs via dice similarity coefficient, and 95% Hausdorff distance, and geometric coverage of clinical CTVs by AID planning target volumes by false positive dice. Coverage of clinical CTVs by AID contour-based plans was evaluated using D100, D95, V100%, and V95%. Results: Between AID and clinical CTVs, the median dice similarity coefficient was 0.66 and the median 95 % Hausdorff distance was 4.0 mm. The median false positive dice of clinical CTV coverage by AID planning target volumes was 0. The median D100 was 1.00, the median D95 was 1.01, the median V100% was 1.00, and the median V95% was 1.00. Increased nodal volume, fraction number, and daily adaptation were associated with reduced clinical CTV coverage by AID-based plans. Conclusion: In one of the first reports on pelvic nodal ART, AID-based plans could adequately cover nodal targets. However, physician review is required due to performance variation. Greater attention is needed for larger, daily-adapted nodes further into treatment.

In Silico Analysis of Adjuvant Head and Neck Online Adaptive Radiation Therapy.

Purpose: Recently developed online adaptive radiation therapy (OnART) systems enable frequent treatment plan adaptation, but data supporting a dosimetric benefit in postoperative head and neck radiation therapy (RT) are sparse. We performed an in silico dosimetric study to assess the potential benefits of a single versus weekly OnART in the treatment of patients with head and neck squamous cell carcinoma in the adjuvant setting. Methods and Materials: Twelve patients receiving conventionally fractionated RT over 6 weeks and 12 patients receiving hypofractionated RT over 3 weeks on a clinical trial were analyzed. The OnART emulator was used to virtually adapt either once midtreatment or weekly based on the patient's routinely performed cone beam computed tomography. The planning target volume (PTV) coverage, dose heterogeneity, and cumulative dose to the organs at risk for these 2 adaptive approaches were compared with the nonadapted plan. Results: In total, 13, 8, and 3 patients had oral cavity, oropharynx, and larynx primaries, respectively. In the conventionally fractionated RT cohort, weekly OnART led to a significant improvement in PTV V100% coverage (6.2%), hot spot (-1.2 Gy), and maximum cord dose (-3.1 Gy), whereas the mean ipsilateral parotid dose increased modestly (1.8 Gy) versus the nonadapted plan. When adapting once midtreatment, PTV coverage improved with a smaller magnitude (0.2%-2.5%), whereas dose increased to the ipsilateral parotid (1.0-1.1 Gy) and mandible (0.2-0.7 Gy). For the hypofractionated RT cohort, similar benefit was observed with weekly OnART, including significant improvement in PTV coverage, hot spot, and maximum cord dose, whereas no consistent dosimetric advantage was seen when adapting once midtreatment. Conclusions: For head and neck squamous cell carcinoma adjuvant RT, there was a limited benefit of single OnART, but weekly adaptations meaningfully improved the dosimetric criteria, predominantly PTV coverage and dose heterogeneity. A prospective study is ongoing to determine the clinical benefit of OnART in this setting.

Optimizing Online Adaptation Timing in the Treatment of Locally Advanced Cervical Cancer.

PURPOSE: Online adaptive radiation therapy (ART) has emerged as a new treatment modality for cervical cancer. Daily online adapting improves target coverage and organ-at-risk (OAR) sparing compared with traditional image guided radiation therapy (IGRT); however, the required resources may not be feasible in a busy clinical setting. Less frequent adapting may still benefit cervical cancer patients due to large volume changes of the uterocervix of the treatment course. In this study, the dosimetry from different online adapt-on-demand schedules was compared. MATERIALS AND METHODS: A retrospective cohort of 10 patients with cervical cancer treated with 260 fractions of definitive daily online ART was included. Plans with different adaptation schedules were simulated with adaptations weekly, every other week, once during treatment, and no adaptations (IGRT). These plans were applied to the synthetic computed tomography (CT) images and contours generated during the patient's delivered daily adaptive workflow. The dosimetry of the weekly replan, every-other-week replan, once replan, and IGRT plans were compared using a paired t test. RESULTS: Compared with traditional IGRT plans, weekly and every-other-week ART plans had similar clinical target volume (CTV) coverage, but statistically significant improved sparing of OARs. Weekly and every-other-week ART had reduced bowel bag V40 by 1.57% and 1.41%, bladder V40 by 3.82% and 1.64%, rectum V40 by 8.49% and 7.50%, and bone marrow Dmean by 0.81% and 0.61%, respectively. Plans with a single adaptation had statistically significantly worse target coverage, and moderate improvements in OAR sparing. Of the 18-dose metrics evaluated, improvements were seen in 15 for weekly ART, 14 for every-other-week ART, and 10 for single ART plans compared with IGRT. When every-other-week ART was compared with weekly ART, both plans had similar CTV coverage and OAR sparing with only small improvements in bone marrow dosimetry with weekly ART. CONCLUSIONS: This retrospective work compares different adapt-on-demand treatment schedules using data collected from patients treated with daily online adaptive radiation therapy. Results suggest weekly or every-other-week online ART is beneficial for reduced OAR dose compared with IGRT by exploiting the gradual changes in the uterocervix target volume.

Long Term Characteristics of Clinical Distribution and Resistance Trends of Carbapenem-Resistant and Extended-Spectrum ß-Lactamase Klebsiella pneumoniae Infections: 2014-2022.

Purpose: Through long-term and large sample size statistical analysis, we revealed the pattern of Klebsiella pneumoniae (KP) infection and drug resistance and provided epidemiological data for the treatment and prevention and control of multidrug-resistant bacterial infection in our hospital. Patients and Methods: Strains were identified using the BD PhoenixTM100 system, minimal inhibitory concentration of antibiotics were determined by the broth method, and data were statistically analyzed using WHONET 5.6 and SPSS27.0. Results: The isolation rate of KP from Enterobacteriaceae (26.2%, 4547/17358) in our hospital showed an increasing annual trend, ranking second only to Escherichia coli. Carbapenem-resistant KP (CRKP) accounted for the highest proportion of carbapenem-resistant Enterobacteriaceae (72.2%, 431/597), showing an upward trend. Infected patients had a male-to-female ratio of approximately 2:1 and were mainly >60 years of age (66.2%), with intensive care units being the most commonly distributed department. Sputum was the most common specimen type (74.0%). Compared with spring and summer, autumn and winter were the main epidemic seasons for KP and extended-spectrum ß-lactamase KP (ESBL-KP). The resistance rate of KP to common antibiotics was low, but all showed an increasing trend each year. ESBL-KP was >90% resistant to piperacillin, amoxicillin/clavulanic acid, and cefotaxime and less resistant to other common antibiotics, but showed an increasing trend in resistance to most antibiotics. CRKP resistance to common antibiotics was high, with resistance rates >90%, excluding amikacin (64.1%), gentamicin (87.4%), cotrimoxazole (44.3%), chloramphenicol (13.6%), and tetracycline (30.5%). Conclusion: KP in our hospital mainly caused pulmonary infection in older men, which occurred frequently in autumn and winter, and the isolation and drug resistance rates showed an increasing trend. Age over 70 years, admission to intensive care unit, and urinary tract infection were found to be the risk factors for CRKP and ESBL-KP-resistance.

Clinical experience on patient-specific quality assurance for CBCT-based online adaptive treatment plan.

PURPOSE: Ethos CBCT-based adaptive radiotherapy (ART) system can generate an online adaptive plan by re-optimizing the initial reference plan based on the patient anatomy at the treatment. The optimization process is fully automated without any room for human intervention. Due to the change in anatomy, the ART plan can be significantly different from the initial plan in terms of plan parameters such as the aperture shapes and number of monitor units (MUs). In this study, we investigated the feasibility of using calculation-based patient specific QA for ART plans in conjunction with measurement-based and calculation-based QA for initial plans to establish an action level for the online ART patient-specific QA. METHODS: A cohort of 98 cases treated on CBCT-based ART system were collected for this study. We performed measurement-based QA using ArcCheck and calculation-based QA using Mobius for both the initial plan and the ART plan for analysis. For online the ART plan, Mobius calculation was conducted prior to the delivery, while ArcCheck measurement was delivered on the same day after the treatment. We first investigated the modulation factors (MFs) and MU numbers of the initial plans and ART plans, respectively. The γ passing rates of initial and ART plan QA were analyzed. Then action limits were derived for QA calculation and measurement for both initial and online ART plans, respectively, from 30 randomly selected patient cases, and were evaluated using the other 68 patient cases. RESULTS: The difference in MF between initial plan and ART-plan was 12.9% ± 12.7% which demonstrates their significant difference in plan parameters. Based on the patient QA results, pre-treatment calculation and measurement results are generally well aligned with ArcCheck measurement results for online ART plans, illustrating their feasibility as an indicator of failure in online ART QA measurements. Furthermore, using 30 randomly selected patient cases, the γ analysis action limit derived for initial plans and ART plans are 89.6% and 90.4% in ArcCheck QA (2%/2 mm) and are 92.4% and 93.6% in Mobius QA(3%/2 mm), respectively. According to the calculated action limits, the ArcCheck measurements for all the initial and ART plans passed QA successfully while the Mobius calculation action limits flagged seven and four failure cases respectively for initial plans and ART plans, respectively. CONCLUSION: An ART plan can be substantially different from the initial plan, and therefore a separate session of ART plan QA is needed to ensure treatment safety and quality. The pre-treatment QA calculation via Mobius can serve as a reliable indicator of failure in online ART plan QA. However, given that Ethos ART system is still relatively new, ArcCheck measurement of initial plan is still in practice. It may be skipped as we gain more experience and have better understanding of the system.

Prevalence of Carbapenem-Resistant Klebsiella pneumoniae Infection in a Northern Province in China: Clinical Characteristics, Drug Resistance, and Geographic Distribution.

PURPOSE: In this study, we analyzed the clinical distribution and drug resistance of carbapenem-resistant Klebsiella pneumoniae (CRKP) strains, the minimum inhibitory concentrations (MIC), MIC50 and MIC90, and geographical distribution in Hebei Province, China. We aimed to provide epidemiological research data, formulate appropriate combined treatment schemes, reasonably select antibiotics, and standardize nosocomial infection control schemes. PATIENTS AND METHODS: A total of 6328 strains of CRKP were collected from 2017 to 2019. The MIC was determined for the drug sensitivity test, and the experimental data were statistically analyzed using WHONET5.6. RESULTS: The detection rate of CRKP increased annually from 13.4% in 2017 to 14.5% in 2018, and 14.6% in 2019. The ratio of males to females was approximately 2:1; 53.6% were elderly, 39% were adults, 4.8% were minors, and 2.5% were newborns. The specimens collected were mainly sputum (70.9%). The resistance rate of CRKP to carbapenems and other ß-lactam antibiotics was found to be increasing, with resistance rates generally greater than 90%. The resistance rate to aminoglycoside antibiotics decreased yearly to approximately 50%, and the resistance rate to quinolones remained unchanged at approximately 80%. From 2017 to 2019, the resistance rate of CRKP in Hebei Province to various antibiotics was high, and the resistance rate to ß-lactam antibiotics increased each year. CONCLUSION: The situation of CRKP resistance is severe in Hebei Province, China. The resistance rate to most antibiotics is very high and shows an upward trend. Among them, the resistance rate to polymyxin is low; however, few resistant strains do exist. MIC50 and MIC90 are higher than their MICs. It mainly causes lung infection in elderly men. This study is helpful to improve the diagnosis, treatment, and prevention of CRKP infection in our province.

Site-agnostic 3D dose distribution prediction with deep learning neural networks.

PURPOSE: Typically, the current dose prediction models are limited to small amounts of data and require retraining for a specific site, often leading to suboptimal performance. We propose a site-agnostic, three-dimensional dose distribution prediction model using deep learning that can leverage data from any treatment site, thus increasing the total data available to train the model. Applying our proposed model to a new target treatment site requires only a brief fine-tuning of the model to the new data and involves no modifications to the model input channels or its parameters. Thus, it can be efficiently adapted to a different treatment site, even with a small training dataset. METHODS: This study uses two separate datasets/treatment sites: data from patients with prostate cancer treated with intensity-modulated radiation therapy (source data), and data from patients with head-and-neck cancer treated with volumetric-modulated arc therapy (target data). We first developed a source model with 3D UNet architecture, trained from random initial weights on the source data. We evaluated the performance of this model on the source data. We then studied the generalizability of the model to the new target dataset via transfer learning. To do this, we built three more models, all with the same 3D UNet architecture: target model, adapted model, and combined model. The source and target models were trained on the source and target data from random initial weights, respectively. The adapted model fine-tuned the source model to the target domain by using the target data. Finally, the combined model was trained from random initial weights on a combined data pool consisting of both target and source datasets. We tested all four models on the target dataset and evaluated quantitative dose-volume histogram metrics for the planning target volume (PTV) and organs at risk (OARs). RESULTS: When tested on the source treatment site, the source model accurately predicted the dose distributions with average (mean, max) absolute dose errors of (0.32%±0.14, 2.37%±0.93) (PTV) relative to the prescription dose, and highest mean dose error of 1.68%±0.76, and highest max dose error of 5.47%± 3.31 for femoral head right. The error in PTV dose coverage prediction is 3.21%±1.51 for D98 , 3.04%±1.69 for D95 , and 1.83%±1.01 for D02 . Averaging across all OARs, the source model predicted the OAR mean dose within 1.38% and the OAR max dose within 3.64%. For the target treatment site, the target model average (mean, max) absolute dose errors relative to the prescription dose for the PTV were (1.08%±0.95, 2.90%±1.35). Left cochlea had the highest mean and max dose errors of 5.37%±5.82 and 8.33%±8.88, respectively. The errors in PTV dose coverage prediction for D98 and D95 were 2.88%±1.59 and 2.55%±1.28, respectively. The target model can predict the OAR mean dose within 2.43% and the OAR max dose within 4.33% on average across all OARs. CONCLUSION: We developed a site-agnostic model for three-dimensional dose prediction and tested its adaptability to a new target treatment site via transfer learning. Our proposed model can make accurate predictions with limited training data.

An Automatic Deep Learning-Based Workflow for Glioblastoma Survival Prediction Using Preoperative Multimodal MR Images: A Feasibility Study.

PURPOSE: Most radiomic studies use the features extracted from the manually drawn tumor contours for classification or survival prediction. However, large interobserver segmentation variations lead to inconsistent features and hence introduce more challenges in constructing robust prediction models. Here, we proposed an automatic workflow for glioblastoma (GBM) survival prediction based on multimodal magnetic resonance (MR) images. METHODS AND MATERIALS: Two hundred eighty-five patients with glioma (210 GBM, 75 low-grade glioma) were included. One hundred sixty-three of the patients with GBM had overall survival data. Every patient had 4 preoperative MR images and manually drawn tumor contours. A 3-dimensional convolutional neural network, VGG-Seg, was trained and validated using 122 patients with glioma for automatic GBM segmentation. The trained VGG-Seg was applied to the remaining 163 patients with GBM to generate their autosegmented tumor contours. The handcrafted and deep learning (DL)-based radiomic features were extracted from the autosegmented contours using explicitly designed algorithms and a pretrained convolutional neural network, respectively. One hundred sixty-three patients with GBM were randomly split into training (n = 122) and testing (n = 41) sets for survival analysis. Cox regression models were trained to construct the handcrafted and DL-based signatures. The prognostic powers of the 2 signatures were evaluated and compared. RESULTS: The VGG-Seg achieved a mean Dice coefficient of 0.86 across 163 patients with GBM for GBM segmentation. The handcrafted signature achieved a C-index of 0.64 (95% confidence interval, 0.55-0.73), whereas the DL-based signature achieved a C-index of 0.67 (95% confidence interval, 0.57-0.77). Unlike the handcrafted signature, the DL-based signature successfully stratified testing patients into 2 prognostically distinct groups. CONCLUSIONS: The VGG-Seg generated accurate GBM contours from 4 MR images. The DL-based signature achieved a numerically higher C-index than the handcrafted signature and significant patient stratification. The proposed automatic workflow demonstrated the potential of improving patient stratification and survival prediction in patients with GBM.

Prospective study of artificial intelligence-based decision support to improve head and neck radiotherapy plan quality.

BACKGROUND AND PURPOSE: Volumetric modulated arc therapy (VMAT) planning for head and neck cancer is a complex process. While the lowest achievable dose for each individual organ-at-risk (OAR) is unknown a priori, artificial intelligence (AI) holds promise as a tool to accurately estimate the expected dose distribution for OARs. We prospectively investigated the benefits of incorporating an AI-based decision support tool (DST) into the clinical workflow to improve OAR sparing. MATERIALS AND METHODS: The DST dose prediction model was based on 276 institutional VMAT plans. Under an IRB-approved prospective trial, the physician first generated a custom OAR directive for 50 consecutive patients (physician directive, PD). The DST then estimated OAR doses (AI directive, AD). For each OAR, the treating physician used the lower directive to form a hybrid directive (HD). The final plan metrics were compared to each directive. A dose difference of 3 Gray (Gy) was considered clinically significant. RESULTS: Compared to the AD and PD, the HD reduced OAR dose objectives by more than 3 Gy in 22% to 75% of cases, depending on OAR. The resulting clinical plan typically met these lower constraints and achieved mean dose reductions between 4.3 and 16 Gy over the PD, and 5.6 to 9.1 Gy over the AD alone. Dose metrics achieved using the HD were significantly better than institutional historical plans for most OARs and NRG constraints for all OARs. CONCLUSIONS: The DST facilitated a significantly improved treatment directive across all OARs for this generalized H&N patient cohort, with neither the AD nor PD alone sufficient to optimally direct planning.

Performance of Deep Learning and Genitourinary Radiologists in Detection of Prostate Cancer Using 3-T Multiparametric Magnetic Resonance Imaging.

BACKGROUND: Several deep learning-based techniques have been developed for prostate cancer (PCa) detection using multiparametric magnetic resonance imaging (mpMRI), but few of them have been rigorously evaluated relative to radiologists' performance or whole-mount histopathology (WMHP). PURPOSE: To compare the performance of a previously proposed deep learning algorithm, FocalNet, and expert radiologists in the detection of PCa on mpMRI with WMHP as the reference. STUDY TYPE: Retrospective, single-center study. SUBJECTS: A total of 553 patients (development cohort: 427 patients; evaluation cohort: 126 patients) who underwent 3-T mpMRI prior to radical prostatectomy from October 2010 to February 2018. FIELD STRENGTH/SEQUENCE: 3-T, T2-weighted imaging and diffusion-weighted imaging. ASSESSMENT: FocalNet was trained on the development cohort to predict PCa locations by detection points, with a confidence value for each point, on the evaluation cohort. Four fellowship-trained genitourinary (GU) radiologists independently evaluated the evaluation cohort to detect suspicious PCa foci, annotate detection point locations, and assign a five-point suspicion score (1: least suspicious, 5: most suspicious) for each annotated detection point. The PCa detection performance of FocalNet and radiologists were evaluated by the lesion detection sensitivity vs. the number of false-positive detections at different thresholds on suspicion scores. Clinically significant lesions: Gleason Group (GG) ≥ 2 or pathological size ≥ 10 mm. Index lesions: the highest GG and the largest pathological size (secondary). STATISTICAL TESTS: Bootstrap hypothesis test for the detection sensitivity between radiologists and FocalNet. RESULTS: For the overall differential detection sensitivity, FocalNet was 5.1% and 4.7% below the radiologists for clinically significant and index lesions, respectively; however, the differences were not statistically significant (P = 0.413 and P = 0.282, respectively). DATA CONCLUSION: FocalNet achieved slightly lower but not statistically significant PCa detection performance compared with GU radiologists. Compared with radiologists, FocalNet demonstrated similar detection performance for a highly sensitive setting (suspicion score ≥ 1) or a highly specific setting (suspicion score = 5), while lower performance in between. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2.

Deep learning-based radiomic features for improving neoadjuvant chemoradiation response prediction in locally advanced rectal cancer.

Radiomic features achieve promising results in cancer diagnosis, treatment response prediction, and survival prediction. Our goal is to compare the handcrafted (explicitly designed) and deep learning (DL)-based radiomic features extracted from pre-treatment diffusion-weighted magnetic resonance images (DWIs) for predicting neoadjuvant chemoradiation treatment (nCRT) response in patients with locally advanced rectal cancer (LARC). 43 Patients receiving nCRT were included. All patients underwent DWIs before nCRT and total mesorectal excision surgery 6-12 weeks after completion of nCRT. Gross tumor volume (GTV) contours were drawn by an experienced radiation oncologist on DWIs. The patient-cohort was split into the responder group (n = 22) and the non-responder group (n = 21) based on the post-nCRT response assessed by postoperative pathology, MRI or colonoscopy. Handcrafted and DL-based features were extracted from the apparent diffusion coefficient (ADC) map of the DWI using conventional computer-aided diagnosis methods and a pre-trained convolution neural network, respectively. Least absolute shrinkage and selection operator (LASSO)-logistic regression models were constructed using extracted features for predicting treatment response. The model performance was evaluated with repeated 20 times stratified 4-fold cross-validation using receiver operating characteristic (ROC) curves and compared using the corrected paired t-test. The model built with handcrafted features achieved the mean area under the ROC curve (AUC) of 0.64, while the one built with DL-based features yielded the mean AUC of 0.73. The corrected paired t-test on AUC showed P-value < 0.05. DL-based features extracted from pre-treatment DWIs achieved significantly better classification performance compared with handcrafted features for predicting nCRT response in patients with LARC.

Joint Prostate Cancer Detection and Gleason Score Prediction in mp-MRI via FocalNet.

Multi-parametric MRI (mp-MRI) is considered the best non-invasive imaging modality for diagnosing prostate cancer (PCa). However, mp-MRI for PCa diagnosis is currently limited by the qualitative or semi-quantitative interpretation criteria, leading to inter-reader variability and a suboptimal ability to assess lesion aggressiveness. Convolutional neural networks (CNNs) are a powerful method to automatically learn the discriminative features for various tasks, including cancer detection. We propose a novel multi-class CNN, FocalNet, to jointly detect PCa lesions and predict their aggressiveness using Gleason score (GS). FocalNet characterizes lesion aggressiveness and fully utilizes distinctive knowledge from mp-MRI. We collected a prostate mp-MRI dataset from 417 patients who underwent 3T mp-MRI exams prior to robotic-assisted laparoscopic prostatectomy. FocalNet was trained and evaluated in this large study cohort with fivefold cross validation. In the free-response receiver operating characteristics (FROC) analysis for lesion detection, FocalNet achieved 89.7% and 87.9% sensitivity for index lesions and clinically significant lesions at one false positive per patient, respectively. For the GS classification, evaluated by the receiver operating characteristics (ROC) analysis, FocalNet received the area under the curve of 0.81 and 0.79 for the classifications of clinically significant PCa (GS ≥ 3 + 4) and PCa with GS ≥ 4 + 3, respectively. With the comparison to the prospective performance of radiologists using the current diagnostic guideline, FocalNet demonstrated comparable detection sensitivity for index lesions and clinically significant lesions, only 3.4% and 1.5% lower than highly experienced radiologists without statistical significance.

Radiofrequency transmit calibration: A multi-center evaluation of vendor-provided radiofrequency transmit mapping methods.

PURPOSE: To determine the accuracy and test-retest repeatability of fast radiofrequency (RF) transmit measurement approaches used in Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI). Spatial variation in the transmitted RF field introduces bias and increased variance in quantitative DCE-MRI metrics including tracer kinetic parameter maps. If unaccounted for, these errors can dominate all other sources of bias and variance. The amount and pattern of variation depend on scanner-specific hardware and software. METHODS: Human tissue mimicking torso and brain phantoms were constructed. RF transmit maps were measured and compared across eight different commercial scanners, from three major vendors, and three clinical sites. Vendor-recommended rapid methods for RF mapping were compared to a slower reference method. Imaging was repeated at all sites after 2 months. Ranges and magnitude of RF inhomogeneity were compared scanner-wise at two time points. Limits of Agreement of vendor-recommended methods and double-angle reference method were assessed. RESULTS: At 3 T, B1 + inhomogeneity spans across 35% in the head and 120% in the torso. Fast vendor provided methods are within 30% agreement with the reference double angle method for both the head and the torso phantom. CONCLUSIONS: If unaccounted for, B1 + inhomogeneity can severely impact tracer-kinetic parameter estimation. Depending on the scanner, fast vendor provided B1 + mapping sequences allow unbiased and reproducible measurements of B1 + inhomogeneity to correct for this source of bias.

Deep transfer learning-based prostate cancer classification using 3 Tesla multi-parametric MRI.

PURPOSE: The purpose of the study was to propose a deep transfer learning (DTL)-based model to distinguish indolent from clinically significant prostate cancer (PCa) lesions and to compare the DTL-based model with a deep learning (DL) model without transfer learning and PIRADS v2 score on 3 Tesla multi-parametric MRI (3T mp-MRI) with whole-mount histopathology (WMHP) validation. METHODS: With IRB approval, 140 patients with 3T mp-MRI and WMHP comprised the study cohort. The DTL-based model was trained on 169 lesions in 110 arbitrarily selected patients and tested on the remaining 47 lesions in 30 patients. We compared the DTL-based model with the same DL model architecture trained from scratch and the classification based on PIRADS v2 score with a threshold of 4 using accuracy, sensitivity, specificity, and area under curve (AUC). Bootstrapping with 2000 resamples was performed to estimate the 95% confidence interval (CI) for AUC. RESULTS: After training on 169 lesions in 110 patients, the AUC of discriminating indolent from clinically significant PCa lesions of the DTL-based model, DL model without transfer learning and PIRADS v2 score ≥ 4 were 0.726 (CI [0.575, 0.876]), 0.687 (CI [0.532, 0.843]), and 0.711 (CI [0.575, 0.847]), respectively, in the testing set. The DTL-based model achieved higher AUC compared to the DL model without transfer learning and PIRADS v2 score ≥ 4 in discriminating clinically significant lesions in the testing set. CONCLUSION: The DeLong test indicated that the DTL-based model achieved comparable AUC compared to the classification based on PIRADS v2 score (p = 0.89).

Repeatability and reproducibility of variable flip angle T1 quantification in the prostate at 3 T.

BACKGROUND: Variable flip angle (VFA) imaging is widely used for the estimation of T1 relaxation in the prostate, but may have limited repeatability and reproducibility due to its sensitivity to B1 + inhomogeneity. PURPOSE: To assess the repeatability and reproducibility of prostate T1 estimation with and without compensating for B1 + variation. STUDY TYPE: Prospective. POPULATION: Twenty-one volunteers were prospectively recruited and scanned twice on two 3 T MRI scanners, resulting in 84 VFA T1 exams. FIELD STRENGTH/SEQUENCE: 3 T/2D saturated turbo fast low angle shot (FLASH) and 3D dual-echo FLASH. ASSESSMENT: Two B1 + mapping techniques, including reference region VFA (RR-VFA) and saturated turbo FLASH (satTFL), were used for B1 + correction, and T1 maps with and without B1 + correction were tested for intrascanner repeatability and interscanner reproducibility. Volumetric regions of interest (ROIs) were drawn on the transition zone, peripheral zone of the prostate, and the obturator internus left and right muscles in the corresponding slices. STATISTICAL TESTS: The average T1 within each ROI for each scan was compared for both intra- and interscanner variability using concordance correlation coefficient and a Bland-Altman plot. RESULTS: Both RR-VFA-corrected T1 and satTFL-corrected T1 showed higher intra- and interscanner correlation (0.89/0.87 and 0.87/0.84, respectively) than VFA T1 (0.84 and 0.74). Bland-Altman plots showed that VFA T1 had wider 95% limits of agreement and a larger range of T1 for each tissue compared with T1 with B1 + correction. DATA CONCLUSION: The application of B1 + correction (both RR-VFA and satTFL) to VFA T1 results in more repeatable and reproducible T1 estimation than VFA T1 . This can potentially provide improved quantification of the prostate dynamic contrast-enhanced MRI parameters. Level of Evidence 1. Technical Efficacy Stage 1. J. Magn. Reson. Imaging 2018.

Prostate DCE-MRI with B 1 + correction using an approximated analytical approach.

PURPOSE: To develop and evaluate a practical B 1 + correction method for prostate dynamic contrast-enhanced (DCE) MRI analysis. THEORY: We proposed a simple analytical B 1 + correction method using a Taylor series approximation to the steady-state spoiled gradient echo signal equation. This approach only requires B 1 + maps and uncorrected pharmacokinetic (PK) parameters as input to estimate the corrected PK parameters. METHODS: The proposed method was evaluated using a prostate digital reference object (DRO), and 82 in vivo prostate DCE-MRI cases. The approximated analytical correction was compared with the ground truth PK parameters in simulation, and compared with the reference numerical correction in in vivo experiments, using percentage error as the metric. RESULTS: The prostate DRO results showed that our approximated analytical approach provided residual error less than 0.4% for both Ktrans and ve , compared to the ground truth. This noise-free residual error was smaller than the noise-induced error using the reference numerical correction, which had a minimum error of 2.1+4.3% with baseline signal-to-noise ratio of 234.5. For the 82 in vivo cases, Ktrans and ve percentage error compared to the reference numerical correction method had a mean of 0.1% (95% central range of [0.0%, 0.2%]) across the prostate volume. CONCLUSION: The approximated analytical B 1 + correction method provides comparable results with less than 0.2% error within 95% central range, compared to reference numerical B 1 + correction. The proposed method is a practical solution for B 1 + correction in prostate DCE-MRI because of its simple implementation.