Search

Background: "Batch effect" in MR images, due to vendor‐specific features, MR machine generations, and imaging parameters, challenges image quality and hinders deep learning (DL) model generalizability. Purpose: We aim to develop a DL model using contrast adjustment and super‐resolution to reduce diffusion‐weighted images (DWIs) diversity across magnetic field strengths and imaging parameters. Study Type: Retrospective. Subjects: The DL model was built using an open dataset from one individual. The MR machine identification model was trained and validated on a dataset of 1134 adults (54% females, 46% males), with 1050 subjects showing no DWI abnormalities and 84 with conditions like stroke and tumors. The 21,000 images were divided into 80% for training, 20% for validation, and 3500 for testing. Field Strength/Sequence: Seven MR scanners from four manufacturers with 1.5 T and 3 T magnetic field strengths. DWIs were acquired using spin‐echo sequences and high‐resolution T2WIs using the T2‐SPACE sequence. Assessment: An experienced, board‐certified radiologist evaluated the effectiveness of restoring high‐resolution T2WI and harmonizing diverse DWI with metrics such as PSNR and SSIM, and the texture and frequency attributes were further analyzed using gray‐level co‐occurrence matrix and 1‐dimensional power spectral density. The model's impact on machine‐specific characteristics was gauged through the performance metrics of a ResNet‐50 model. Comprehensive statistical tests were employed for statistical robustness, including McNemar's test and the Dice index. Results: Our DL protocol reduced DWI contrast and resolution variation. ResNet‐50 model's accuracy decreased from 0.9443 to 0.5786, precision from 0.9442 to 0.6494, recall from 0.9443 to 0.5786, and F1 score from 0.9438 to 0.5587. The t‐SNE visualization indicated more consistent image features across multiple MR devices. Autoencoder halved learning iterations; Dice coefficient >0.74 confirmed signal reproducibility in 84 lesions. Conclusion: This study presents a DL strategy to mitigate batch effects in diffusion MR images, improving their quality and generalizability. Evidence Level: 3 Technical Efficacy: Stage 1 [ABSTRACT FROM AUTHOR]

Background and Objectives: Integrating large language models (LLMs) such as GPT-4 Turbo into diagnostic imaging faces a significant challenge, with current misdiagnosis rates ranging from 30–50%. This study evaluates how prompt engineering and confidence thresholds can improve diagnostic accuracy in neuroradiology. Methods: We analyze 751 neuroradiology cases from the American Journal of Neuroradiology using GPT-4 Turbo with customized prompts to improve diagnostic precision. Results: Initially, GPT-4 Turbo achieved a baseline diagnostic accuracy of 55.1%. By reformatting responses to list five diagnostic candidates and applying a 90% confidence threshold, the highest precision of the diagnosis increased to 72.9%, with the candidate list providing the correct diagnosis at 85.9%, reducing the misdiagnosis rate to 14.1%. However, this threshold reduced the number of cases that responded. Conclusions: Strategic prompt engineering and high confidence thresholds significantly reduce misdiagnoses and improve the precision of the LLM diagnostic in neuroradiology. More research is needed to optimize these approaches for broader clinical implementation, balancing accuracy and utility. [ABSTRACT FROM AUTHOR]

Terahertz quantum cascade laser (QCL) sources based on intra-cavity difference frequency generation are currently the only electrically pumped monolithic semiconductor light sources operating at room temperature in the 1–6-THz spectral range. Relying on the active regions with the giant second-order nonlinear susceptibility and the Cherenkov phase-matching scheme, these devices demonstrated drastic improvements in performance in the past several years and can now produce narrow-linewidth single-mode terahertz emission that is tunable from 1 to 6 THz with power output sufficient for imaging and spectroscopic applications. This paper reviews the progress of this technology. Recent efforts in wave function engineering using a new active region design based on a dual-upper-state concept led to a significant enhancement of the optical nonlinearity of the active region for efficient terahertz generation. The transfer of Cherenkov devices from their native semi-insulating InP substrates to high-resistivity silicon substrates resulted in a dramatic improvement in the outcoupling efficiency of terahertz radiation. Cherenkov terahertz QCL sources based on the dual-upper-state design have also been shown to exhibit ultra-broadband comb-like terahertz emission spectra with more than one octave of terahertz frequency span. The broadband terahertz QCL sources operating in continuous-wave mode produces the narrow inter-mode beat-note linewidth of 287 Hz, which indicates frequency comb operation of mid-infrared pumps and thus supports potential terahertz comb operation. Finally, we report the high-quality terahertz imaging obtained by a THz imaging system using terahertz QCL sources based on intra-cavity difference frequency generation.

In plant populations, limited seed dispersal may generate the aggregation of related individuals, which is referred to as spatial genetic structure (SGS). If short-distance pollination frequently occurs in populations with SGS, it may lead to frequent mating between related individuals (biparental inbreeding). In this study, we examined patterns of pollen dispersal and inbreeding under such conditions in a population of the wind-pollinated tree species Fagus crenata Blume using microsatellite analyses of adult trees and dispersed seeds. The population showed weak but significant SGS and the pollen dispersal in the population involved a combination of distance-dependent (largely short-range) dispersion at the local scale and long-distance transport. Although SGS and distance-dependent pollination co-occurred, individual inbreeding coefficients for both adults and seeds were almost zero, suggesting that mating generating sound seeds was genetically random with little or no inbreeding, probably because of the weak SGS, long-distance pollen dispersal, and postpollination mechanisms including self-incompatibility and inbreeding depression. The production of sound seeds through genetically random mating may be important for maintaining F. crenata populations because the resulting seedlings may be less affected by inbreeding depression than those resulting from the mating of related individuals. Key words: dispersal kernel, Fagus crenata, inbreeding coefficient, outcrossing rate, paternity analysis, wind pollination. Resume: Dans les populations vegetales, la dispersion limitee des graines peut generer une agregation des individus apparentes, phenomene que l'on appelle structure spatiale genetique (SSG). Si la pollinisation sur de courtes distances survient frequemment dans les populations avec SSG, ceci peut mener a un croisement frequent entre des individus apparentes (endogamie biparentale). Dans cette etude, les auteurs ont examine les patrons de dispersion du pollen et l'endogamie sous de telles conditions chez une population de Fagus crenata Blume, une espece d'arbres a pollinisation anemophile, a l'aide d'analyses de microsatellites des arbres adultes et des graines dispersees. La population presentait une SSG faible mais significative, et la dispersion du pollen dans la population impliquait la combinaison d'une dispersion en fonction de la distance (principalement de courte distance) a l'echelle locale et d'un transport de longue distance. Meme si la SSG et la pollinisation en fonction de la distance survenaient conjointement, les coefficients individuels d'endogamie des adultes et des graines etaient presque de zero, suggerant que le croisement generant des graines saines etait genetiquement aleatoire avec peu ou pas d'endogamie, probablement a cause de la faible SSG, de la dispersion du pollen sur une longue distance et des mecanismes post-pollinisation incluant l'auto-incompatibilite et la depression endogamique. La production de graines saines au moyen du croisement genetiquement aleatoire peut etre importante pour maintenir les populations de F. crenata car les semis resultants peuvent etre moins affectes par la depression endogamique que ceux qui resultent du croisement d'individus apparentes. [Traduit par la Redaction] Mots-cles : dispersion des grains, Fagus crenata, coefficient d'endogamie, taux de croisement eloigne, analyse de paternite, pollinisation anemophile., Introduction Gene dispersal via seeds and pollen strongly influences the genetic diversity and structure of plant populations. Genetic differentiation among populations is reduced by gene dispersal (i.e., gene flow) even [...]