Pilot Lightweight Denoising Algorithm for Multiple Sclerosis on Spine MRI.

Multiple sclerosis (MS) is a severely debilitating disease which requires accurate and timely diagnosis. MRI is the primary diagnostic vehicle; however, it is susceptible to noise and artifact which can limit diagnostic accuracy. A myriad of denoising algorithms have been developed over the years for medical imaging yet the models continue to become more complex. We developed a lightweight algorithm which utilizes the image's inherent noise via dictionary learning to improve image quality without high computational complexity or pretraining through a process known as orthogonal matching pursuit (OMP). Our algorithm is compared to existing traditional denoising algorithms to evaluate performance on real noise that would commonly be encountered in a clinical setting. Fifty patients with a history of MS who received 1.5 T MRI of the spine between the years of 2018 and 2022 were retrospectively identified in accordance with local IRB policies. Native resolution 5 mm sagittal images were selected from T2 weighted sequences for evaluation using various denoising techniques including our proposed OMP denoising algorithm. Peak signal to noise ratio (PSNR) and structural similarity index (SSIM) were measured. While wavelet denoising demonstrated an expected higher PSNR than other models, its SSIM was variable and consistently underperformed its comparators (0.94 ± 0.10). Our pilot OMP denoising algorithm provided superior performance with greater consistency in terms of SSIM (0.99 ± 0.01) with similar PSNR to non-local means filtering (NLM), both of which were superior to other comparators (OMP 37.6 ± 2.2, NLM 38.0 ± 1.8). The superior performance of our OMP denoising algorithm in comparison to traditional models is promising for clinical utility. Given its individualized and lightweight approach, implementation into PACS may be more easily incorporated. It is our hope that this technology will provide improved diagnostic accuracy and workflow optimization for Neurologists and Radiologists, as well as improved patient outcomes.

Time-Dependent Deep Learning Prediction of Multiple Sclerosis Disability.

The majority of deep learning models in medical image analysis concentrate on single snapshot timepoint circumstances, such as the identification of current pathology on a given image or volume. This is often in contrast to the diagnostic methodology in radiology where presumed pathologic findings are correlated to prior studies and subsequent changes over time. For multiple sclerosis (MS), the current body of literature describes various forms of lesion segmentation with few studies analyzing disability progression over time. For the purpose of longitudinal time-dependent analysis, we propose a combinatorial analysis of a video vision transformer (ViViT) benchmarked against traditional recurrent neural network of Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) architectures and a hybrid Vision Transformer-LSTM (ViT-LSTM) to predict long-term disability based upon the Extended Disability Severity Score (EDSS). The patient cohort was procured from a two-site institution with 703 patients' multisequence, contrast-enhanced MRIs of the cervical spine between the years 2002 and 2023. Following a competitive performance analysis, a VGG-16-based CNN-LSTM was compared to ViViT with an ablation analysis to determine time-dependency of the models. The VGG16-LSTM predicted trinary classification of EDSS score in 6 years with 0.74 AUC versus the ViViT with 0.84 AUC (p-value < 0.001 per 5 × 2 cross-validation F-test) on an 80:20 hold-out testing split. However, the VGG16-LSTM outperformed ViViT when patients with only 2 years of MRIs (n = 94) (0.75 AUC versus 0.72 AUC, respectively). Exact EDSS classification was investigated for both models using both classification and regression strategies but showed collectively worse performance. Our experimental results demonstrate the ability of time-dependent deep learning models to predict disability in MS using trinary stratification of disability, mimicking clinical practice. Further work includes external validation and subsequent observational clinical trials.

Presurgical Upgrade Prediction of DCIS to Invasive Ductal Carcinoma Using Time-dependent Deep Learning Models with DCE MRI.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To determine whether time-dependent deep learning models can outperform single timepoint models in predicting preoperative upgrade of ductal carcinoma in situ (DCIS) to invasive malignancy on dynamic contrastenhanced (DCE) breast MRI without lesion segmentation prerequisite. Materials and Methods In this exploratory study, 154 cases of biopsy-proven DCIS (25 upgraded at surgery and 129 not upgraded) were selected consecutively from a retrospective cohort of preoperative DCE MRI in women with an average age of 58.6 years at time of diagnosis from 2012 to 2022. Binary classification was implemented with convolutional neural network-long short-term memory (CNN-LSTM) architectures benchmarked against traditional CNNs without manual segmentation of the lesions. Combinatorial performance analysis of ResNet50 versus VGG16-based models was performed with each contrast phase. Binary classification area under the receiver operating characteristic curve (AUC) was reported. Results VGG16-based models consistently provided better hold-out test AUCs than ResNet50 in CNN and CNNLSTM studies (multiphase test AUC: 0.67 versus 0.59, respectively, for CNN models; P = .04 and 0.73 versus 0.62 for CNN-LSTM models; P = .008). The time-dependent model (CNN-LSTM) provided a better multiphase test AUC over single-timepoint (CNN) models (0.73 versus 0.67, P = .04). Conclusion Compared with single-timepoint architectures, sequential deep learning algorithms using preoperative DCE MRI improved prediction of DCIS lesions upgraded to invasive malignancy without the need for lesion segmentation. ©RSNA, 2024.

The 14-3-3 proteins of Arabidopsis regulate root growth and chloroplast development as components of the photosensory system.

The 14-3-3 proteins specifically bind a number of client proteins to influence important pathways, including flowering timing via the photosensory system. For instance, 14-3-3 proteins influence the photosensory system through interactions with Constans (CO) protein. 14-3-3 associations with the photosensory system were further studied in this investigation using 14-3-3 T-DNA insertion mutants to study root and chloroplast development. The 14-3-3 µ T-DNA insertion mutant, 14-3-3µ-1, had shorter roots than the wild type and the difference in root length could be influenced by light intensity. The 14-3-3 ν T-DNA insertion mutants also had shorter roots, but only when grown under narrow-bandwidth red light. Five-day-old 14-3-3 T-DNA insertion and co mutants all had increased root greening compared with the wild type, which was influenced by light wavelength and intensity. However, beyond 10 d of growth, 14-3-3µ-1 roots did not increase in greening as much as wild-type roots. This study reveals new developmental roles of 14-3-3 proteins in roots and chloroplasts, probably via association with the photosensory system.

Activator protein 1 (AP-1) contributes to EpCAM-dependent breast cancer invasion.

INTRODUCTION: EpCAM is a cell-surface glycoprotein that is overexpressed in the majority of epithelial carcinomas. However, the functional role of EpCAM in regulating cancer invasion remains controversial, and the mechanism(s) underlying EpCAM-mediated regulation of breast cancer invasion remain to be defined. METHODS: EpCAM expression was manipulated in breast cancer cell lines using RNA interference and cDNA expression constructs. Recombinant EpCAM was used to rescue EpCAM signaling following specific ablation of EpCAM. Protein and gene expression, invasion, transcription factor activity, and protein phosphorylation were measured using standard molecular biology techniques. RESULTS: In loss-of-function, and gain-of-function experiments we demonstrate that EpCAM expression is associated with increased breast cancer invasion in vitro and in vivo. We demonstrate further that specific ablation of EpCAM expression is associated with decreased activator protein-1 (AP-1) transcription factor activity. Phosphoprotein analyses confirm that specific ablation of EpCAM is associated with decreased phosphorylation of the AP-1 subunit c-Jun. Recombinant soluble extracellular EpCAM (rEpCAM) is able to rescue invasion, AP-1 transcription factor activity, and c-Jun phosphorylation in a dose-dependent fashion. Pharmacologic inhibitors, and constitutively active constructs of the c-Jun N-terminal kinase (JNK) signal transduction pathway, suggest that the impact of EpCAM expression on AP-1 transcription factor activity is mediated through the JNK pathway. In functional rescue experiments, forced expression of c-Jun rescues invasion in breast cancer cells following specific ablation of EpCAM. CONCLUSIONS: These data demonstrate for the first time that EpCAM expression can influence the JNK/AP-1 signal transduction pathway, and suggest that modulation of AP-1 transcription factor activity contributes to EpCAM-dependent breast cancer invasion. These data have important implications for the design and application of molecular therapies targeting EpCAM.

Cetuximab-associated pulmonary toxicity in concurrent chemoradiation for the treatment of a squamous cell carcinoma of the head and neck.

BACKGROUND: Cetuximab is a common EGFR monoclonal antibody used with radiotherapy to treat head-and-neck cancer. Severe pulmonary toxicity, including interstitial lung disease (ILD), caused by cetuximab is rare. METHODS: We describe a patient who developed ILD and acute respiratory failure after concurrent chemoradiation with cetuximab for oropharyngeal squamous cell carcinoma, and review the literature. RESULTS: A patient developed acute respiratory failure 2 months after starting concurrent chemoradiation with cetuximab and was hospitalized in intensive care after a procedure for progressive respiratory distress. Cultures and serology were negative for infection and radiologic findings were consistent with drug associated pneumonitits. Steroids were administered until the patient was stabilized. The patient fully recovered 1 month after the onset of respiratory distress, although he died of recurrent disease 10 months after completing treatment. CONCLUSION: Although severe pulmonary toxicity caused by EGFR inhibitors has been well described in the literature, ILD caused by cetuximab, an EGFR monoclonal antibody, is rare and not well-documented. Given its life-threatening effects, awareness of this potential side effect and early diagnosis is critical.

Activation Mechanism of Oncogenic Deletion Mutations in BRAF, EGFR, and HER2.

Activating mutations in protein kinases drive many cancers. While how recurring point mutations affect kinase activity has been described, the effect of in-frame deletions is not well understood. We show that oncogenic deletions within the ß3-αC loop of HER2 and BRAF are analogous to the recurrent EGFR exon 19 deletions. We identify pancreatic carcinomas with BRAF deletions mutually exclusive with KRAS mutations. Crystal structures of BRAF deletions reveal the truncated loop restrains αC in an active "in" conformation, imparting resistance to inhibitors like vemurafenib that bind the αC "out" conformation. Characterization of loop length explains the prevalence of five amino acid deletions in BRAF, EGFR, and HER2 and highlights the importance of this region for kinase activity and inhibitor efficacy.

Parabolic flight induces changes in gene expression patterns in Arabidopsis thaliana.

Our primary objective was to evaluate gene expression changes in Arabidopsis thaliana in response to parabolic flight as part of a comprehensive approach to the molecular biology of spaceflight-related adaptations. In addition, we wished to establish parabolic flight as a tractable operations platform for molecular biology studies. In a succession of experiments on NASA's KC-135 and C-9 parabolic aircraft, Arabidopsis plants were presented with replicated exposure to parabolic flight. Transcriptome profiling revealed that parabolic flight caused changes in gene expression patterns that stood the statistical tests of replication on three different flight days. The earliest response, after 20 parabolas, was characterized by a prominence of genes associated with signal transduction. After 40 parabolas, this prominence was largely replaced by genes associated with biotic and abiotic stimuli and stress. Among these responses, three metabolic processes stand out in particular: the induction of auxin metabolism and signaling, the differential expression of genes associated with calcium-mediated signaling, and the repression of genes associated with disease resistance and cell wall biochemistry. Many, but not all, of these responses are known to be involved in gravity sensing in plants. Changes in auxin-related gene expression were also recorded by reporter genes tuned to auxin signal pathways. These data demonstrate that the parabolic flight environment is appropriate for molecular biology research involving the transition to microgravity, in that with replication, proper controls, and analyses, gene expression changes can be observed in the time frames of typical parabolic flight experiments.

Transcriptional repression of epithelial cell adhesion molecule contributes to p53 control of breast cancer invasion.

p53 is a tumor suppressor gene with well-characterized roles in cell cycle regulation, apoptosis, and maintenance of genome stability. Recent evidence suggests that p53 may also contribute to the regulation of migration and invasion. Epithelial cell adhesion molecule (EpCAM) is a transmembrane glycoprotein that is overexpressed in the majority of human epithelial carcinomas, including breast and colorectal carcinomas. We show by chromatin immunoprecipitation assays that p53 interacts with a candidate p53 binding site within the EpCAM gene. p53-mediated transcriptional repression of EpCAM was confirmed in gain-of-function and loss-of-function experimental systems. Induction of wild-type p53 was associated with a significant dose-dependent decrease in EpCAM expression; conversely, specific ablation of p53 was associated with a significant increase in EpCAM expression. At the functional level, specific ablation of p53 expression is associated with increased breast cancer invasion, and this effect is abrogated by concomitant specific ablation of EpCAM expression. Taken together, these biochemical and functional data are the first demonstration that (a) wild-type p53 protein binds to a response element within the EpCAM gene and negatively regulates EpCAM expression, and (b) transcriptional repression of EpCAM contributes to p53 control of breast cancer invasion.

14-3-3 isoforms participate in red light signaling and photoperiodic flowering.

Members of the 14-3-3 family of proteins participate in signal transduction by modulating flux through various pathways. Potential subfunctionalization within this family has produced a suite of related proteins with diverse client interactions and discrete localization. The associated study assesses the biological roles of two specific 14-3-3 isoforms, using genetic, biochemical and physiological assays to ascertain potential nodes of interaction. Arabidopsis T-DNA insertion mutants representing the nu and mu isoforms exhibited a short, yet clear delay in flowering time on long days. Tests of hypocotyl growth inhibition under narrow bandwidth light indicated a hyposensitivity to red light, while responses to blue and far-red light were normal. These physiological tests suggest a mechanistic link between 14-3-3 proteins, red light sensing, and the pathways that control photoperiodic flowering. The precise entry point into the pathway was assessed using yeast two hybrid assays targeted against specific proteins active in the circadian oscillator, light transduction and photoperiodic flowering. Yeast two hybrid interaction was observed with CONSTANS (CO), and then confirmed with coimmunoprecipitation. Functional interaction with phyB leading to defects in flowering time and direct interaction with CONSTANS circumstantially places these specific 14-3-3 isoforms into the pathway that regulates the transition between vegetative and floral development.

The 14-3-3 Proteins mu and upsilon influence transition to flowering and early phytochrome response.

14-3-3 proteins regulate a diverse set of biological responses but developmental phenotypes associated with 14-3-3 mutations have not been described in plants. Here, physiological and biochemical tests demonstrate interactions between 14-3-3s and the well-established mechanisms that govern light sensing and photoperiodic flowering control. Plants featuring homozygous disruption of 14-3-3 isoforms upsilon and mu display defects in light sensing and/or response. Mutant plants flower late and exhibit long hypocotyls under red light, with little effect under blue or far-red light. The long hypocotyl phenotype is consistent with a role for 14-3-3 upsilon and mu in phytochrome B signaling. Yeast two-hybrid and coimmunoprecipitation assays indicate that 14-3-3 upsilon and mu proteins physically interact with CONSTANS, a central regulator of the photoperiod pathway. Together, these data indicate a potential role for specific 14-3-3 isoforms in affecting photoperiodic flowering via interaction with CONSTANS, possibly as integrators of light signals sensed through the phytochrome system.