Airborne Wind Energy (AWE) is looking very promising for harnessing high- altitute winds and aiding in the transition from fossil fuels to sustainable en- ergy. The ground-based kite system in plays a crucial role in autonomously estimating tether force, which depends on various factors such as wind speed, the kite’s orientation relative to the wind vector in its figure-eight trajectory and Latitute as well as Longitude. To predict tether force, we have em- ployed testing of four regression machine learning models which have shown merit in similar fields. The machine learning models which were tested upon were: Linear Regression Support Vector Machine Regression Random Forest Regression XGBoost Regressor Gradient Boosting After getting the metrics which were Mean Absolute Error(MAE), Root Mean Square Error(RMSE) and R 2 Error, we concluded that XGBoost Re- gressor gave us the best metrics in all three categories.

Summary: A rare case of bilateral perineal hernia was reported in a 24-year-old Italian Trotter mare. This condition is described as a protrusion of retroperitoneal fat, and/or the abdominal or pelvic viscera through the pelvic diaphragm which supports the rectal wall. It is commonly observed in olduncastrated old male dogs, but occurs very rarely in large ruminants; to the Authors’ knowledge, it has never been described in horses. Chronic weight loss is the main clinical sign, even if the main concern regards the entrapment and the strangulation of the bowel content. Reaching a diagnosis involves clinical signs, palpation, ultrasound evaluation and rectal examination. The approach chosen in all the cases of perineal hernia described in large ruminants is appositional herniorrhaphy; for this reason, it was utilised as the first surgical choice in the present case. Due to its failure, a new surgical approach was used The second procedure involved the transposition of the semimembranosus muscle. The short-term outcome after the second surgery presented no recurrence or complications. It is therefore necessary to consider perineal hernia among the various differential diagnosis of perineal swellings in horses. Additional cases are required to determine its benefits.

Our current planetary crisis moves the need for effective ecosystem restoration centerstage and compels us to explore unusual options. We here propose exploring combinatorial approaches to restoration practices: management practices are drawn at random and combined from a locally relevant pool of possible management interventions, thus creating an experimental gradient in the number of interventions. This will move the current degree of interventions to higher dimensionality, opening new opportunities for unlocking unknown synergistic effects. In this concept, regional restoration hubs play an important role as guardians of locally relevant information and sites of experimental exploration. Data collected from such studies could feed into a global database, which could be used to learn about general principles of combined restoration practices, helping to refine future experiments. Such combinatorial approaches to exploring restoration intervention options may be our best hope yet to achieve decisive progress in ecological restoration at the timescale needed to mitigate and reverse the most severe losses.

A document by منى عبدالله محمد الخاوي. Click on the document to view its contents.

Necrotizing pancreatitis in a patient diagnosed with SLE; rare case reportMaryam Javid1 ,Tohid Damideh2, Omid pourbagherian1 , Mehdi Jafarpour31 Immunology Research Center, Tabriz University of Medical Sciences, Tabriz, Iran2Faculty of Medicine, Tabriz University of Medical Sciences, Tabriz, Iran3Connective Tissue Diseases Research Center, Tabriz University of Medical Sciences, Tabriz, IranDrafting of the manuscript: Maryam Javid, Mehdi JafarpourData Collection: Omid pourbagherian, Tohid DamidehCritical revision of the manuscript for important intellectual content: Mehdi Jafarpour, Maryam JavidCase supervision: Mehdi JafarpourCorresponding author: Mehdi JafarpourAddress: Connective Tissue Diseases Research Center, Flat 1, Imam Reza Hospital, Tabriz, IranPhone number: 0098 41 33332704Fax: 0098 41 35413520Mobile: 09148651269Corresponding author e-mail address: jafarpourmehdi1360@gmail.comKeywords: Necrotizing pancreatitis, Systemic Lupus Erythematosus, inflammation Written informed permission from the patient to publish the facts and photographs was granted.

Article Type:A case report

Biochemical changes in the cervix during labor are not well understood, in part because of a lack of technology capable of safely probing the pregnant cervix in vivo. FT-IR spectroscopy has the potential to address these needs because it is a non-invasive optical technique that can sensitively detect changes in biochemical components. A total of 30 pregnant participants undergoing either spontaneous or induced labor were recruited. We detected several biochemical changes during labor, including a significant decrease in FT-IR spectral features associated with collagen and other extracellular matrix proteins attributed to collagen dispersion, an increase in spectral features associated with blood, and in features indicative of lipid-based molecules. Our results have demonstrated that FT-IR spectroscopy is sensitive to multiple biochemical remodeling changes in the cervix during labor. FT-IR spectroscopy may be a valuable noninvasive tool for objective cervical assessment to potentially guide clinical labor management.

The rhizosphere influence on the soil microbiome and function of crop wild progenitors remains virtually unknown, despite its relevance to develop microbiome-oriented tools in sustainable agriculture. Here, we quantified the rhizosphere influence -- a comparison between rhizosphere and bulk soil samples -- on bacterial, fungal, protists and invertebrates communities and on soil multifunctionality across nine crop wild progenitors in their sites of origin. Overall, rhizosphere influence was higher on abundant taxa across the four microbial groups, and had a positive influence on increased rhizosphere carbon storage and nutrient contents compared to bulk soils. The rhizosphere influence on abundant soil microbiomes were more important for soil multifunctionaility than rare taxa and envirommental conditions. Our results are a starting point to uncover the roles of both abundant and rare soil taxa in enhancing multifunctionality in agroecosystems.

The introduction of YOLOv9, the latest version of the You Only Look Once (YOLO) series, has led to its widespread adoption across various scenarios. This paper is the first to apply the YOLOv9 algorithm model to the fracture detection task as computer-assisted diagnosis (CAD) to help radiologists and surgeons to interpret X-ray images. Specifically, this paper trained the model on the GRAZPEDWRI-DX dataset and extended the training set using data augmentation techniques to improve the model performance. Experimental results demonstrate that compared to the mAP 50-95 of the current state-of-the-art (SOTA) model, the YOLOv9 model increased the value from 42.16% to 43.73%, with an improvement of 3.7%. The implementation code is publicly available at https://github.com/RuiyangJu/YOLOv9-Fracture-Detection.

Based on morphological and molecular evidence, we herein described Cremanthodium dibangii (Asteraceae) as a new species from India. The species was found growing in the subalpine meadows of Dibang Valley, Arunachal Pradesh, India. A Nucleotide Basic Local Alignment Search Tool (BLASTn) search of the nrDNA ITS region and the plastid trnC-petN intergenic spacer sequences revealed sequence similarities to C. stenoglossum Y.Ling & S.W.Liu (98.88 % identity) and C. rhodocephalum Diels (99.62 % identity), respectively. Further, phylogenetic tree reconstruction of the ITS placed C. dibangii as sister to C. stenoglossum and C. campanulatam, while the trnC-petN tree depicted C. dibangii as a separate clade. Morphologically it is related to C. thomsonii C.B.Clarke and C. pulchrum R.D.Good in having reniform leaves and ligulate heads but differs from these in having size of basal leaves size; number of stem leaves, ray florets and phyllaries; and length of style. The diagnostic characteristics, affinities with allied taxa and colored illustration are provided to facilitate easy identification.

A document by Dinberu Oyamo. Click on the document to view its contents.

Atypical hemolytic uremic syndrome (aHUS) is an infrequently encountered complement-mediated thrombotic microangiopathy (TMA) usually associated with germline variants in genes of the complement system. Clinical findings of microangiopathic hemolytic anemia, thrombocytopenia, and acute kidney injury (AKI) with severe hypertension arise due to aberrant complement protein activation in the circulation and significant endothelial damage. Transplant-associated thrombotic microangiopathy has been increasingly recognized after high dose carboplatin, etoposide, and melphalan-chemotherapy followed by autologous hematopoietic stem cell rescue for treatment of children with neuroblastoma (NB). We report the case of a 13-month-old boy with metastatic neuroblastoma who developed aHUS during the first cycle of induction chemotherapy. Germline testing revealed a Complement factor H (CFH) gene mutation, Cys357Arg, which is currently classified as a variant of uncertain significance (VUS), although likely pathogenic based on molecular modeling as well as this patient’s clinical presentation. The patient has been successfully managed with complement blockade therapy with no recurrence of disease. We review presentations of neuroblastoma with hypertension, along with AKI and thrombocytopenia, to raise awareness about the potential for aHUS in patients with newly diagnosed NB.

Colorectal cancer (CRC) is a common and highly metastatic cancer affecting people worldwide. Drug resistance and unwanted side effects are some of the limitations of current treatments for CRC. Naringenin (NAR) is a naturally occurring compound found in abundance in various citrus fruits such as oranges, grapefruits, and tomatoes. It possesses a diverse range of pharmacological and biological properties that are beneficial for human health. Numerous studies have highlighted its antioxidant, anti-cancer, and anti-inflammatory activities, making it a subject of interest in scientific research. This review provides a comprehensive overview of the effects of NAR on CRC. The study’s findings indicated that NAR: 1) interacts with estrogen receptors, 2) regulates the expression of genes related to the p53 signaling pathway, 3) promotes apoptosis by increasing the expression of proapoptotic genes (Bax, caspase9, and p53) and downregulation of the antiapoptotic gene Bcl2, 4) inhibits the activity of enzymes involved in cell survival and proliferation, 5) decreases cyclin D1 levels, 6) reduces the expression of cyclin-dependent kinases (Cdk4, Cdk6, Cdk7) and anti-apoptotic genes (Bcl2, x-IAP, c-IAP-2) in CRC cells. In vitro CDK2 binding assay was also performed, showing that the NAR derivatives had better inhibitory activities on CDK2 than NAR. Based on the findings of this study, NAR is a potential therapeutic agent for CRC. Additional pharmacology and pharmacokinetics studies are required to fully elucidate the mechanisms of action of NAR and establish the most suitable dose for subsequent clinical investigations.

Objective: To explore implementation of two intrapartum trials with compelling findings: BUMPES (position in second stage of labour in nulliparous women with epidural), and RESPITE (remifentanil intravenous patient-controlled analgesia). Design: Qualitative interview study. Setting: UK NHS Trusts and Universities. Sample: Purposively sampled investigators from RESPITE and BUMPES trials and clinicians providing intrapartum care: midwives, anaesthetists and obstetricians recruited using existing networks and snowball sampling. Methods: Semi-structured virtual interviews. Thematic analysis was underpinned by Capability Opportunity Motivation Behaviour Change Framework. Results: Twenty-nine interview participants across 19 maternity units: 11 clinical academics, 10 midwives, 4 obstetricians, 4 anaesthetists. Most (25/29) were aware of one or both trials. BUMPES had been implemented in 4/19 units (one original trial site) and RESPITE in 3/19 units (two trial sites). Access to sufficient resources, training, exposure to interventions, support from leaders, and post-trial dissemination and implementation activities all facilitated uptake of interventions. Some clinicians were opposed to the intervention or disagreed with trial conclusions. However competing priorities in terms of staff time and a plethora of initiatives in maternity care, emerged as key barrier to implementation. Conclusions: Compelling trial findings were not implemented widely, and numerous barriers and facilitators were identified. Large-scale improvement programmes and evidence-based national guidelines may mean single trials have limited potential to change practice. There is a need to examine how intervention implementation is prioritised to optimise safety outcomes in the context of workforce restrictions, limited resources and large arrays of competing priorities including statutory requirements, that have increased in maternity care.

Objective The aim of the study was to describe the characteristics of otherwise healthy children with obstructive sleep apnea (OSA; OSA-I) and children with OSA and obesity (OSA-II) treated with long term continuous positive airway pressure (CPAP) or noninvasive ventilation (NIV) in 2019 in France. Design National cross-sectional survey. Patients Children with OSA-I and OSA-II. Main outcome measures Initiation criteria, age, adherence, equipment and settings Results Patients with OSA-I and OSA-II represented 6% (n=84, 71% males) and 10% (n=144, 72% males) of the national cohort, respectively. The apnea-hypopnea index (63% vs 76%), alone or combined with nocturnal gas exchange (25% vs 21%, for OSA-II and OSA-I patients respectively) were used as initiation criteria of CPAP/NIV. OSA-II patients were older at CPAP/NIV initiation (mean age 11.0±4.0 vs 6.8±4.5 years, p<0.001) and were treated for a longer time (2.3±2.6 vs 1.3±1.5 years, p=0.008) than OSA-I patients. NIV was used in 6% of OSA-I patients and 13% of OSA-II patients (p=0.142). Nasal mask was the most used interface in both groups. Mean CPAP level was higher in OSA-II patients as compared to OSA-I patients (8.7±2.0 vs 7.7±2.4 cmH 2O, p=0.02). Objective compliance was comparable (mean use 6.8±2.6 vs 5.9±3.0 hours/night in OSA-I and OSA-II, respectively, p=0.054). Conclusion Six and 10% of children treated with long term CPAP/NIV in France in 2019 had OSA-I and OSA-II, respectively. Both groups were preferentially treated with CPAP and were comparable except for age, with OSA-II patients being older.

Shelter-building insects are important ecosystem engineers, playing critical roles in structuring arthropod communities. Nonetheless, the influence of leaf shelters and arthropods on plant-associated microbiota remains largely unexplored. Arthropods that visit or inhabit plants can contribute to the leaf microbial community, resulting in significant changes in plant-microbe interactions. By artificially constructing leaf shelters, we provide evidence that shelter-building insects influence not only the arthropod community structure but also impact the phyllosphere microbiota. Leaf shelters exhibited higher abundance and richness of arthropods, leading to changes in the associated arthropod community composition. These shelters also altered the composition and community structure of phyllosphere microbiota, promoting greater richness and diversity of bacteria at the phyllosphere. In leaf shelters, microbial diversity increased with the richness and diversity of herbivores. These findings demonstrate the critical role of leaf shelters in structuring arthropod communities by facilitating colonization and influencing microbial communities through altered microhabitats and community interactions.

Background Prenatal echocardiographic assessment of fetal cardiac function has become increasingly important. Fetal two-dimensional speckle-tracking echocardiography (2D-STE) allows the determination of global and segmental functional cardiac parameters. Prenatal diagnostics is relying increasingly on artificial intelligence, whose algorithms transform the way clinicians use ultrasound in their daily workflow. The purpose of this study was to demonstrate the feasibility of whether less experienced operators can handle and might benefit from an automated tool of 2D-STE in the clinical routine. Methods A total of 136 unselected, normal, singleton, second- and third-trimester fetuses with normofrequent heart rates were examined by targeted ultrasound. 2D-STE was performed separately by beginner and expert semiautomatically using a GE Voluson E10 (FetalHQ®, GE Healthcare, Chicago, IL, USA). Several fetal cardiac parameters were calculated (end-diastolic diameter (ED), sphericity index (SI), global longitudinal strain (EndoGLS), fractional shortening (FS)) and assigned to gestational age (GA). Bland-Altman plots were used to test agreement between both operators. Results The mean maternal age was 33 years, and the mean maternal body mass index prior to pregnancy was 24.78 kg/m 2. The GA ranged from 16.4 to 32.0 weeks (average 22.9 weeks). Averaged endoGLS value of the beginner was -18.57 % ± 6.59 percentage points (pp) for the right and -19.58 % ± 5.63 pp for the left ventricle, that of the expert -14.33 % ± 4.88 pp and -16.37 % ± 5.42 pp. With increasing GA, right ventricular endoGLS decreased slightly while the left ventricular were almost constant. The statistical analysis for endoGLS showed a Bland-Altman-Bias of -4.24 pp ± 8.06 pp for the right and -3.21 pp ± 7.11 pp for the left ventricle. The Bland-Altman-Bias of the ED in both ventricles in all analyzed segments ranged from -0.49 mm ± 1.54 mm to -0.10 mm ± 1.28 mm, that for FS from -0.33 pp ± 11.82 pp to 3.91 pp ± 15.56 pp and that for SI from -0.38 ± 0.68 to -0.15 ± 0.45. Conclusions Between both operators, our data indicated that 2D-STE analysis showed excellent agreement for cardiac morphometry parameters (ED and SI), and good agreement for cardiac function parameters (EndoGLS and FS). Due to its complexity, the application of fetal 2D-STE remains the domain of scientific-academic perinatal ultrasound and should be placed preferably in the hands of skilled operators. At present, from our perspective, an implementation into clinical practice ‘on-the-fly’ cannot be recommended.

Introductions of quantum communications, enabled by advancements in quantum computing, is expected to play a significant role in the field of communications. Inherent properties of quantum objects, such as superposition and entanglement have the potential to provide novel solutions to overcome the challenges encountered by classical communication systems in bandwidth-intensive applications such as media transmission. This research explores the performance of a quantum communication system in image transmission using quantum superposition and investigates its performance using a simple quantum channel model. With increase of channel noise, there are significant gains in the rate distortion performance of images transmitted over the quantum channel, compared to an ideal classical channel. This novel attempt in constructing a quantum communication-based image transmission system indicates the potential of the approach to be applied to satisfy the ever-increasing demands of high-quality media transmission applications.

In Chlamydomonas, the directly light-gated, plasma membrane-localized (PM) cation channels channelrhodopsins ChR1 and ChR2 are the primary photoreceptors for phototaxis. Their targeting and abundance is essential for optimal movement responses. However, our knowledge how Chlamydomonas achieves this is still at its infancy. Here we show that ChR1 internalization occurs via light-stimulated endocytosis. Prior or during endocytosis ChR1 is modified and forms high molecular mass complexes. These are the solely detectable ChR1 forms in extracellular vesicles (EVs) and their abundance therein dynamically changes upon illumination. The ChR1-containing EVs are secreted via the PM and/or the ciliary base. In line with this, ciliogenesis mutants exhibit increased ChR1 degradation rates. Further, we establish involvement of two cysteine proteases in its turnover: CEP1, a papain-type C1A member, and a calpain. ΔCEP1 knock-out strains lack light-induced ChR1 degradation, whereas ChR2 degradation was unaffected. Low light stimulates CEP1 expression, which is regulated via phototropin, a SPA1 E3 ubiquitin ligase and cAMP. Further, mutant and inhibitor analyses revealed involvement of the small GTPase ARL11 and SUMOylation in ChR1 targeting to the eyespot and cilia. Our study thus defines the degradation pathway of this central photoreceptor of Chlamydomonas and identifies novel elements involved in its homeostasis and targeting.

Diarrhea is the most prevalent side effect of afatinib as an EGFR tyrosine kinase inhibitor. The current study is to investigate the potential protective mechanisms of crofelemer or loperamide in an animal model of afatinib-induced diarrhea. Rats were randomized as the control, afatinib (50 mg/kg), afatinib plus loperamide (50 mg/kg) or afatinib plus crofelemer (50 mg/kg) group. Rats received drugs or saline through oral gavage daily for consecutive 7 days. Diarrhea, weight, serum biomarkers, gut histology or ultrastructure and multiomic changes were analyzed daily or at day 8. Afatinib induced significant diarrhea, weight loss, elevated serum levels of endotoxin, IL-6, IL-1β and TNF-α in rats. Mucosal damage was most prominent in distal ileum, showing edema, inflammatory infiltration, epithelial villus atrophy or fusion and loss of tight junction. Both loperamide and crofelemer conferred protection against afatinib-induced diarrhea and gut damage. Transcriptomic enrichment analysis showed that PPAR and IL-17 signaling pathway are among the top modified pathways in the ileum and colon of the afatinib group, respectively. Metabolomic profiling identified 318 differently abundant metabolites when comparing the afatinib and the control groups, with the most prominent enriched metabolic pathways being metabolism of xenobiotics by cytochrome P450, retinol metabolism and lysine degradation. Afatinib significantly decreased microbial diversity, which is not fully restored by administration of crofelemer or loperamide. Correlation analysis showed that cecal microbiota were significantly correlated with metabolite profiles. Loperamide and crofelemer attenuate afatinib-induced diarrhea and intestinal damage in rats, possibly through regulating microbiota-metabolic axis.

To the Editor, Blood hypereosinophilia (HE), defined as eosinophils in peripheral blood over 1.5 G/L, is clinical challenge in modern medicine [1]. A variety of causes and symptoms make patients undergo many tests before final diagnosis is made and the necessary therapy is initiated [1]. Factors predicting the final diagnosis might prove useful, as some of the more costly and invasive tests could be skipped [1]. The aim of the study was to find prognostic factors which may identify the cause of HE. Medical records of 39 patients with observed blood HE detected in at least one test who visited the Department of Allergology in Gdansk in the years 2010–2022 were analysed. In general, we supplemented the standard tests with additional ones when predicting neoplastic or autoimmune diseases. The peripheral blood morphology, serum biochemistry, parasitological examination, immunological studies (antineutrophil cytoplasmic antibodies (ANCA) test), additional tests, including the aspirin provocation test (performed in 46% of patients (n=18)), spirometry, histopathology and cytology of the bone marrow (done in 56% of patients (n=22)) were performed. The statistical analysis was conducted using binary logistic regression performed using Statistica 13 software.

This study successfully synthesized lamellar Zn2Ti3O8, TiO2, and rod-shaped ZnO catalysts through hydrothermal and calcination methods, comparing the catalytic performance of the three catalysts on MgH2. Notably, Zn2Ti3O8 demonstrated significant synergistic enhancement in the dehydrogenation temperature and desorption kinetics of MgH2. Experimental results revealed that the MgH2 + 12 wt% Zn2Ti3O8 composite material commenced dehydrogenation at approximately 185 °C, around 160 °C lower than pure MgH2. Furthermore, at 325 °C, MgH2 release d only 3.06 wt% H2 within 20 min, whereas the Zn2Ti3O8 doped composite system achieved the same hydrogen release at 225 °C within 7 min. After complete dehydrogenation, the MgH2 + 12 wt% Zn2Ti3O8 composite material-initiated hydrogen absorption at 40 °C and absorbed approximately 6.05 wt% H2 within 60 min at 200 °C. The activation energies for hydrogen absorption and desorption of the composite material decreased by 22.25 kJ·mol-1 and 44.67 kJ·mol-1, respectively, compared to pure MgH2. Cycling experiments demonstrated that after 20 cycles, the MgH2 + 12wt% Zn2Ti3O8 composite material maintained excellent cyclic stability, with hydrogen storage capacity still exceeding 95%. Hence, this composite material exhibits outstanding catalytic and hydrogen storage performance, holding potential application value in enhancing hydrogen release rates and reducing activation energy.

Objective The objective of the study was to investigate the differences in the gut microbiomes of neonates delivered via cesarean section compared to those born by vaginal delivery, and to identify the predominant microbial taxa present in each group. Study design A prospective observational study. Setting At Her Royal Highness Maha Chakri Sirindhorn Medical Center, Srinakharinwirot University, Thailand. Sample Fecal sample obtained by 281 healthy neonates born between February 2021 and April 2023. The study population was divided into two groups: 139 neonates born via vaginal delivery and 142 neonates born via cesarean section. Methods The microbiota composition of each neonate’s fecal sample was identified by using 16S ribosomal ribonucleic acid metagenomic sequencing. Main Outcome Measures Neonatal gut microbiome abundancy and diversity was identified according to route of delivery. Results Neonates delivered vaginally exhibited a gut microbiome with higher abundance and diversity than those delivered by cesarean delivery. Bifidobacterium was the dominant genus in both groups. Bifidobacterium breve was the dominant species and was significantly higher in cesarean-delivered neonates compared to those delivered vaginally (24.0% and 9.2%, respectively) (p<0.0001). However, the taxonomy of only 89 (64.0%) and 44 (31.43%) fecal samples could be identified from the vaginal and cesarean delivery groups, respectively. Conclusions Route of delivery is associated with neonatal gut microbiome abundancy and diversity. Neonates delivered via vaginal delivery exhibited higher diversity but lower abundancy of the dominant species in the gut microbiome.

1. IntroductionMagnetic resonance imaging (MRI) plays a key role in the diagnosis and management of a range of neurological conditions (Atlas, 2009). However, the growing demand for brain MRI examinations, along with a global shortage of radiologists, is taking its toll on healthcare systems. Increasingly, radiologists are unable to fulfill their reporting requirements within contracted hours, leading to substantial reporting delays (NHS, 2021)(Wood et al., 2021). Concerns about fatigue-related diagnostic errors are also mounting as radiologists become increasingly overworked (Vosshenrich et al., 2021). Ultimately, reporting delays and errors lead to delays in treatment; for many abnormalities, this results in poorer patient outcomes and inflated healthcare costs (Adams et al., 2005).Potentially, artificial intelligence (AI) could be used to relieve some of the pressure on radiology departments, for example by supporting real-time triaging of examinations (Annarumma et al., 2019)(Yala et al., 2019)(Wood et al., 2022)(Verburg et al., 2022)(Agarwal et al., 2023)(Booth et al., 2023)(Agarwal et al., 2023) or assisting radiologists to reduce errors in radiology reports. To date, efforts in this direction have largely relied on deep learning models trained on expert-labelled datasets (Gulshan, 2016))(Titano et al., 2018)(De Fauw et al., 2018)(Ardila et al., 2019)(McKinney et al., 2020)(Wood et al., 2019)(Wood et al., 2022)(Din et al., 2023)(Chelliah et al., 2024). However, there are key limitations to this approach. First, the growing pressure on clinical services has made it increasingly difficult to justify using radiologists’ time to manually annotate images for research purposes; obtaining large, clinically representative training datasets therefore represents a bottleneck to model development (Wood et al., 2020)(Benger et al., 2023)(Wood et al., 2024). Second, the use of categorically labelled datasets in conjunction with supervised learning methods inherently restricts classification to a pre-defined, fixed set of classes. As such, whenever a new classification task emerges, additional labelled training examples are needed. This poses a considerable problem for neuroradiological applications, where the dynamic nature of clinical demands constantly alters the landscape of automation possibilities. For example, the class of ‘tumours’ may become insufficient for a detection task when there is a new demand for a particular type of tumour; additional labelling of the particular type of tumour is required (Louis et al., 2021).These issues, among others, have led to a growing interest in multi-modal (e.g., text-vision) self-supervised methods which enable computer vision models to learn directly from free-text radiology reports (Zhang et al., 2022)(Boecking et al., 2022)(Bannur et al., 2023). Radiology reports represent promising training data since they i) contain detailed descriptions and impressions of all image findings observed by expert radiologists; and ii) are typically stored alongside imaging data on hospital picture archiving and communication systems (PACS) and so are relatively easy to obtain. To date, however, the application of self-supervised methods has largely been limited to image recognition tasks involving chest radiographs - due in part to the availability of open-access, paired image-text datasets such as MIMIC Chest X-ray (MIMIC-CXR)  (Johnson et al., 2019)(Agarwal et al., 2024). To our knowledge there has been no previous demonstration of text-vision models for either brain abnormality detection or for the highly complex modality of MRI (Wood et al., 2022).Here, we present a self-supervised text-vision framework which learns to detect clinically relevant abnormalities from unlabelled hospital brain MRI scans. Our two-step training approach proceeded as follows. First, a dedicated neuroradiological language model - NeuroBERT - was trained to generate fixed-dimensional vector representations of neuroradiology reports via domain-specific self-supervised learning tasks. Next, convolutional neural networks (CNN) - one per MRI sequence type, covering the full range of sequences performed during routine examinations - learnt to map individual brain scans to their corresponding text vector representations by optimising a mean square error (MSE) loss. Once trained, our text-vision framework can be used to detect abnormalities in unreported brain MRI examinations by scoring scans against suitable query sentences (e.g., ‘this is a normal study’, or ‘there is an acute stroke’ etc.), opening a range of classification-based applications including automated triage (Fig. 1), diagnosis, and treatment response assessment. Potentially, our framework could also operate as a clinical decision support tool by suggesting findings to radiologists, detecting errors in provisional reports, and retrieving and displaying examples of pathologies from historical examinations that could be relevant to the current case based on textual descriptors.