Vestibular schwannomas (VS) are benign tumors that can lead to hearing loss, balance issues, and tinnitus. Gamma Knife Radiosurgery (GKS) is a common treatment for VS, aimed at halting tumor growth and preserving neurological function. Accurate monitoring of VS volume before and after GKS is essential for assessing treatment efficacy. To evaluate the accuracy of an artificial intelligence (AI) algorithm, originally developed to identify NF2-SWN-related VS, in segmenting non-NF2-SWN-related VS and detecting volume changes pre- and post-GKS. We hypothesize this AI algorithm, trained on NF2-SWN-related VS data, will accurately apply to non-NF2-SWN VS and VS treated with GKS. In this retrospective cohort study, we reviewed data from an established Gamma Knife database, identifying 16 patients who underwent GKS for VS and had pre- and post-GKS scans. Contrast-enhanced T1-weighted MRI scans were analyzed with both manual segmentation and the AI algorithm. DICE similarity coefficients were computed to compare AI and manual segmentations, and a paired t-test was used to assess statistical significance. Volume changes for pre- and post-GKS scans were calculated for both segmentation methods. The mean DICE score between AI and manual segmentations was 0.91 (range 0.79-0.97). Pre- and post-GKS DICE scores were 0.91 (range 0.79-0.97) and 0.92 (range 0.81-0.97), indicating high spatial overlap. AI-segmented VS volumes pre- and post-GKS were consistent with manual measurements, with high DICE scores indicating strong spatial overlap. The AI algorithm processed scans within 5 min, suggesting it offers a reliable, efficient alternative for clinical monitoring. DICE scores showed high similarity between manual and AI segmentations. The pre- and post-GKS VS volume percentage changes were also similar between manual and AI-segmented VS volumes, indicating that our AI algorithm can accurately detect changes in tumor growth.

The present study aimed to develop and validate a fusion model based on multi-phase contrast-enhanced computed tomography (CECT) radiomics features combined with clinical features to preoperatively predict the expression levels of Ki-67 in patients with gastric cancer (GC). A total of 164 patients with GC who underwent surgical treatment at our hospital between September 2015 and September 2023 were retrospectively included and were randomly divided into a training set (n=114) and a testing set (n=50). Using Pyradiomics, radiomics features were extracted from multi-phase CECT images and were combined with significant clinical features through various machine learning algorithms [support vector machine (SVM), random forest (RandomForest), K-nearest neighbors (KNN), LightGBM and XGBoost] to build a fusion model. Receiver operating characteristic, area under the curve (AUC), calibration curve and decision curve analysis (DCA) were used to evaluate, validate and compare the predictive performance and clinical utility of the model. Among the three single-phase models, for the arterial phase model, the SVM radiomics model had the highest AUC value in the training set, which was 0.697; and the RandomForest radiomics model had the highest AUC value in the testing set, which was 0.658. For the venous phase model, the SVM radiomics model had the highest AUC value in the training set, which was 0.783; and the LightGBM radiomics model had the highest AUC value in the testing set, which was 0.747. For the delayed phase model, the KNN radiomics model had the highest AUC value in the training set, which was 0.772; and the SVM radiomics model had the highest AUC in the testing set, which was 0.719. The clinical feature model had the lowest AUC values in both the training set and the testing set, which were 0.614 and 0.520, respectively. Notably, the multi-phase model and the fusion model, which were constructed by combining the clinical features and the multi-phase features, demonstrated excellent discriminative performance, with the fusion model achieving AUC values of 0.933 and 0.817 in the training and testing sets, thus outperforming other models (DeLong test, both P<0.05). The calibration curve showed that the fusion model had goodness of fit (Hosmer-Lemeshow test, >0.5 in the training and validation sets). The DCA showed that the net benefit of the fusion model in identifying high expression of Ki-67 was improved compared with that of other models. Furthermore, the fusion model achieved an AUC value of 0.805 in the external validation data from The Cancer Imaging Archive. In conclusion, the fusion model established in the present study was revealed to have excellent performance and is expected to serve as a non-invasive tool for predicting Ki-67 status and guiding clinical treatment.

The US healthcare system faces significant challenges, including clinician burnout, operational inefficiencies, and concerns about patient safety. Artificial intelligence (AI), particularly generative AI, has the potential to address these challenges, but its adoption, effectiveness, and barriers to implementation are not well understood. To evaluate the current state of AI adoption in US healthcare systems, assess successes and barriers to implementation during the early generative AI era. This cross-sectional survey was conducted in Fall 2024, and included 67 health systems members of the Scottsdale Institute, a collaborative of US non-profit healthcare organizations. Forty-three health systems completed the survey (64% response rate). Respondents provided data on the deployment status and perceived success of 37 AI use cases across 10 categories. The primary outcomes were the extent of AI use case development, piloting, or deployment, the degree of reported success for AI use cases, and the most significant barriers to adoption. Across the 43 responding health systems, AI adoption and perceptions of success varied significantly. Ambient Notes, a generative AI tool for clinical documentation, was the only use case with 100% of respondents reporting adoption activities, and 53% reported a high degree of success with using AI for Clinical Documentation. Imaging and radiology emerged as the most widely deployed clinical AI use case, with 90% of organizations reporting at least partial deployment, although successes with diagnostic use cases were limited. Similarly, many organizations have deployed AI for clinical risk stratification such as early sepsis detection, but only 38% report high success in this area. Immature AI tools were identified a significant barrier to adoption, cited by 77% of respondents, followed by financial concerns (47%) and regulatory uncertainty (40%). Ambient Notes is rapidly advancing in US healthcare systems and demonstrating early success. Other AI use cases show varying degrees of adoption and success, constrained by barriers such as immature AI tools, financial concerns, and regulatory uncertainty. Addressing these challenges through robust evaluations, shared strategies, and governance models will be essential to ensure effective integration and adoption of AI into healthcare practice.

Undersampling accelerates signal acquisition at the expense of introducing artifacts. Removing these artifacts is a fundamental problem in signal processing and this task is also called signal reconstruction. Through modeling signals as the superimposed exponential functions, deep learning has achieved fast and high-fidelity signal reconstruction by training a mapping from the undersampled exponentials to the fully sampled ones. However, the mismatch, such as undersampling rates (25 % vs. 50 %), anatomical region (knee vs. brain), and contrast configurations (PDw vs. T₂w), between the training and target data will heavily compromise the reconstruction. To overcome this limitation, we propose Alternating Deep Low-Rank (ADLR), which combines deep learning solvers and classic optimization solvers. Experimental validation on the reconstruction of synthetic and real-world biomedical magnetic resonance signals demonstrates that ADLR can effectively alleviate the mismatch issue and achieve lower reconstruction errors than state-of-the-art methods.

<b>Objective:</b>To development a deep learning（DL） model based on conventional MRI for automatic segmentation and differential diagnosis of nasopharyngeal carcinoma（NPC） and nasopharyngeal lymphoma（NPL）. <b>Methods:</b>The retrospective study included 142 patients with NPL and 292 patients with NPC who underwent conventional MRI at Renmin Hospital of Wuhan University from June 2012 to February 2023. MRI from 80 patients were manually segmented to train the segmentation model. The automatically segmented regions of interest（ROIs） formed four datasets: T1 weighted images（T1WI）, T2 weighted images（T2WI）, T1 weighted contrast-enhanced images（T1CE）, and a combination of T1WI and T2WI. The ImageNet-pretrained ResNet101 model was fine-tuned for the classification task. Statistical analysis was conducted using SPSS 22.0. The Dice coefficient loss was used to evaluate performance of segmentation task. Diagnostic performance was assessed using receiver operating characteristic（ROC） curves. Gradient-weighted class activation mapping（Grad-CAM） was imported to visualize the model's function. <b>Results:</b>The DICE score of the segmentation model reached 0.876 in the testing set. The AUC values of classification models in testing set were as follows: T1WI: 0.78（95%<i>CI</i> 0.67-0.81）, T2WI: 0.75（95%<i>CI</i> 0.72-0.86）, T1CE: 0.84（95%<i>CI</i> 0.76-0.87）, and T1WI+T2WI: 0.93（95%<i>CI</i> 0.85-0.94）. The AUC values for the two clinicians were 0.77（95%<i>CI</i> 0.72-0.82） for the junior, and 0.84（95%<i>CI</i> 0.80-0.89） for the senior. Grad-CAM analysis revealed that the central region of the tumor was highly correlated with the model's classification decisions, while the correlation was lower in the peripheral regions. <b>Conclusion:</b>The deep learning model performed well in differentiating NPC from NPL based on conventional MRI. The T1WI+T2WI combination model exhibited the best performance. The model can assist in the early diagnosis of NPC and NPL, facilitating timely and standardized treatment, which may improve patient prognosis.

To assess the image quality and biventricular function utilizing a free-breathing artificial intelligence cine method with motion correction (FB AI MOCO). A total of 72 participants (mean age 38.3 ± 15.4 years, 40 males) prospectively enrolled in this single-center, cross-sectional study underwent cine scans using standard breath-holding (BH) cine and FB AI MOCO cine at 3.0 Tesla. The image quality of the cine images was evaluated with a 5-point Ordinal Likert scale based on blood-pool to myocardium contrast, endocardial edge definition, and artifacts, and overall quality score was calculated by the equal weight average of all three criteria, apparent signal to noise ratio (aSNR), estimated contrast to noise ratio (eCNR) were assessed. Biventricular functional parameters including Left Ventricular (LV), Right Ventricular (RV) End-Diastolic Volume (EDV), End-Systolic Volume (ESV), Stroke Volume (SV), Ejection Fraction (EF), and LV End-Diastolic Mass (LVEDM) were also assessed. Comparison between two sequences was assessed using paired t-test and Wilcoxon signed-rank test, correlation using Pearson correlation. The agreement of quantitative parameters was assessed using intraclass correlation coefficient (ICC) and Bland-Altman analysis. P < 0.05 was statistically significant. The total acquisition time of the entire stack for FB AI MOCO cine (14.7 s ± 1.9 s) was notably shorter than that for standard BH cine (82.6 s ± 11.9 s, P < 0.001). The aSNR between FB AI MOCO cine and standard BH cine has no significantly difference (76.7 ± 20.7 vs. 79.8 ± 20.7, P = 0.193). The eCNR of FB AI MOCO cine was higher than standard BH cine (191.6 ± 54.0 vs. 155.8 ± 68.4, P < 0.001), as was the scores of blood-pool to myocardium contrast (4.6 ± 0.5 vs. 4.4 ± 0.6, P = 0.003). Qualitative scores including endocardial edge definition (4.2 ± 0.5 vs. 4.3 ± 0.7, P = 0.123), artifact presence (4.3 ± 0.6 vs. 4.1 ± 0.8, P = 0.085), and overall image quality (4.4 ± 0.4 vs. 4.3 ± 0.6, P = 0.448), showed no significant differences between the two methods. Representative RV and LV functional parameters - including RVEDV (102.2 (86.4, 120.4) ml vs. 104.0 (88.5, 120.3) ml, P = 0.294), RVEF (31.0 ± 11.1 % vs. 31.2 ± 11.0 %, P = 0.570), and LVEDV (106.2 (86.7, 131.3) ml vs. 105.8 (84.4, 130.3) ml, P = 0.450) - also did not differ significantly between the two methods. Strong correlations (r > 0.900) and excellent agreement (ICC > 0.900) were found for all biventricular functional parameters between the two sequences. In subgroups with reduced LVEF (<50 %, n = 24) or elevated heart rate (≥80  bpm, n = 17), no significant differences were observed in any biventricular functional metrics (P > 0.05 for all) between the two sequences. In comparison to multiple BH cine, the FB AI MOCO cine achieved comparable image quality and biventricular functional parameters with shorter scan times, suggesting its promising potential for clinical applications.

Confounding factors inherent in medical images can significantly impact the causal exploration capabilities of deep learning models, resulting in compromised accuracy and diminished generalization performance. In this paper, we present an innovative methodology named CausalMixNet that employs query-mixed intra-attention and key&value-mixed inter-attention to probe causal relationships between input images and labels. For mitigating unobservable confounding factors, CausalMixNet integrates the non-local reasoning module (NLRM) and the key&value-mixed inter-attention (KVMIA) to conduct a front-door adjustment strategy. Furthermore, CausalMixNet incorporates a patch-masked ranking module (PMRM) and query-mixed intra-attention (QMIA) to enhance mediator learning, thereby facilitating causal intervention. The patch mixing mechanism applied to query/(key&value) features within QMIA and KVMIA specifically targets lesion-related feature enhancement and the inference of average causal effect inference. CausalMixNet consistently outperforms existing methods, achieving superior accuracy and F1-scores across in-domain and out-of-domain scenarios on multiple datasets, with an average improvement of 3% over the closest competitor. Demonstrating robustness against noise, gender bias, and attribute bias, CausalMixNet excels in handling unobservable confounders, maintaining stable performance even in challenging conditions.

Chemical exchange saturation transfer (CEST) MRI is a powerful molecular imaging technique for detecting metabolites through proton exchange. While CEST MRI provides high sensitivity, its clinical application is hindered by prolonged scan time due to the need for imaging across numerous frequency offsets for parameter estimation. Since scan time is directly proportional to the number of frequency offsets, identifying and selecting the most informative frequency can significantly reduce acquisition time. We propose a novel deep learning-based framework that integrates frequency selection and parameter estimation to accelerate CEST MRI. Our method leverages channel pruning via batch normalization to identify the most informative frequency offsets while simultaneously training the network for accurate parametric map prediction. Using data from six healthy volunteers, channel pruning selects 13 informative frequency offsets out of 53 without compromising map quality. Images from selected frequency offsets were reconstructed using the MR Multitasking method, which employs a low-rank tensor model to enable under-sampling of k-space lines for each frequency offset, further reducing scan time. Predicted parametric maps of amide proton transfer (APT), nuclear overhauser effect (NOE), and magnetization transfer (MT) based on these selected frequencies were comparable in quality to maps generated using all frequency offsets, achieving superior performance compared to Fisher information-based selection methods from our previous work. This integrated approach has the potential to reduce the whole-brain CEST MRI scan time from the original 5:30 min to under 1:30 min without compromising map quality. By leveraging deep learning for frequency selection and parametric map prediction, the proposed framework demonstrates its potential for efficient and practical clinical implementation. Future studies will focus on extending this method to patient populations and addressing challenges such as B₀ inhomogeneity and abnormal signal variation in diseased tissues.

Amide Proton Transfer (APT) technique is a novel functional MRI technique that enables quantification of protein metabolism, but its wide application is largely limited in clinical settings by its long acquisition time. One way to reduce the scanning time is to obtain fewer frequency offset images during image acquisition. However, sparse frequency offset images are not inadequate to fit the z-spectral, a curve essential to quantifying the APT effect, which might compromise its quantification. In our study, we develop a deep learning-based model that allows for reconstructing dense frequency offsets from sparse ones, potentially reducing scanning time. We propose to leverage time-series convolution to extract both short and long-range spatial and frequency features of the APT imaging sequence. Our proposed model outperforms other seq2seq models, achieving superior reconstruction with a peak signal-to-noise ratio of 45.8 (95% confidence interval (CI): [44.9 46.7]), and a structural similarity index of 0.989 (95% CI:[0.987 0.993]) for the tumor region. We have integrated a weighted layer into our model to evaluate the impact of individual frequency offset on the reconstruction process. The weights assigned to the frequency offset at ±6.5 ppm, 0 ppm, and 3.5 ppm demonstrate higher significance as learned by the model. Experimental results demonstrate that our proposed model effectively reconstructs dense frequency offsets (n = 29, from 7 to -7 with 0.5 ppm as an interval) from data with 21 frequency offsets, reducing scanning time by 25%. This work presents a method for shortening the APT imaging acquisition time, offering potential guidance for parameter settings in APT imaging and serving as a valuable reference for clinicians.

To compare clinical ratings and signal-to-noise ratio (SNR) measures of a commercially available Deep Learning-based MRI reconstruction method (T2_(DR)) against conventional T2- turbo spin echo brain MRI (T2_(CN)). 100 consecutive patients with various neurological conditions underwent both T2_(DR) and T2_(CN) on a Siemens Vida 3 T scanner with a 64-channel head coil in the same examination. Acquisition times were 3.33 min for T2_(CN) and 1.04 min for T2_(DR). Four neuroradiologists evaluated overall image quality (OIQ), diagnostic safety (DS), and image artifacts (IA), blinded to the acquisition mode. SNR and SNR_eff (adjusted for acquisition time) were calculated for air, grey- and white matter, and cerebrospinal fluid. The mean patient age was 43.6 years (SD 20.3), with 54 females. The distribution of non-diagnostic ratings did not differ significantly between T2_(CN) and T2_(DR) (IA p = 0.108; OIQ: p = 0.700 and DS: p = 0.652). However, when considering the full spectrum of ratings, significant differences favouring T2_(CN) emerged in OIQ (p = 0.003) and IA (p < 0.001). T2_(CN) had higher SNR (157.9, SD 123.4) than T2_(DR) (112.8, SD 82.7), p < 0.001, but T2_(DR) demonstrated superior SNR_eff (14.1, SD 10.3) compared to T2_(CN) (10.8, SD 8.5), p < 0.001. Our results suggest that while T2_(DR) may be clinically applicable for a diagnostic setting, it does not fully match the quality of high-standard conventional T2_(CN), MRI acquisitions.