Kinetic magnetism in triangular moiré materials.

Magnetic properties of materials ranging from conventional ferromagnetic metals to strongly correlated materials such as cuprates originate from Coulomb exchange interactions. The existence of alternate mechanisms for magnetism that could naturally facilitate electrical control has been discussed theoretically1-7, but an experimental demonstration8 in an extended system has been missing. Here we investigate MoSe2/WS2 van der Waals heterostructures in the vicinity of Mott insulator states of electrons forming a frustrated triangular lattice and observe direct evidence of magnetic correlations originating from a kinetic mechanism. By directly measuring electronic magnetization through the strength of the polarization-selective attractive polaron resonance9,10, we find that when the Mott state is electron-doped, the system exhibits ferromagnetic correlations in agreement with the Nagaoka mechanism.

A repeating fast radio burst associated with a persistent radio source.

The dispersive sweep of fast radio bursts (FRBs) has been used to probe the ionized baryon content of the intergalactic medium1, which is assumed to dominate the total extragalactic dispersion. Although the host-galaxy contributions to the dispersion measure appear to be small for most FRBs2, in at least one case there is evidence for an extreme magneto-ionic local environment3,4 and a compact persistent radio source5. Here we report the detection and localization of the repeating FRB 20190520B, which is co-located with a compact, persistent radio source and associated with a dwarf host galaxy of high specific-star-formation rate at a redshift of 0.241 ± 0.001. The estimated host-galaxy dispersion measure of approximately [Formula: see text] parsecs per cubic centimetre, which is nearly an order of magnitude higher than the average of FRB host galaxies2,6, far exceeds the dispersion-measure contribution of the intergalactic medium. Caution is thus warranted in inferring redshifts for FRBs without accurate host-galaxy identifications.

A fast radio burst source at a complex magnetized site in a barred galaxy.

Fast radio bursts (FRBs) are highly dispersed, millisecond-duration radio bursts1-3. Recent observations of a Galactic FRB4-8 suggest that at least some FRBs originate from magnetars, but the origin of cosmological FRBs is still not settled. Here we report the detection of 1,863 bursts in 82 h over 54 days from the repeating source FRB 20201124A (ref. 9). These observations show irregular short-time variation of the Faraday rotation measure (RM), which scrutinizes the density-weighted line-of-sight magnetic field strength, of individual bursts during the first 36 days, followed by a constant RM. We detected circular polarization in more than half of the burst sample, including one burst reaching a high fractional circular polarization of 75%. Oscillations in fractional linear and circular polarizations, as well as polarization angle as a function of wavelength, were detected. All of these features provide evidence for a complicated, dynamically evolving, magnetized immediate environment within about an astronomical unit (AU; Earth-Sun distance) of the source. Our optical observations of its Milky-Way-sized, metal-rich host galaxy10-12 show a barred spiral, with the FRB source residing in a low-stellar-density interarm region at an intermediate galactocentric distance. This environment is inconsistent with a young magnetar engine formed during an extreme explosion of a massive star that resulted in a long gamma-ray burst or superluminous supernova.

A constraint on historic growth in global photosynthesis due to increasing CO2.

The global terrestrial carbon sink is increasing1-3, offsetting roughly a third of anthropogenic CO2 released into the atmosphere each decade1, and thus serving to slow4 the growth of atmospheric CO2. It has been suggested that a CO2-induced long-term increase in global photosynthesis, a process known as CO2 fertilization, is responsible for a large proportion of the current terrestrial carbon sink4-7. The estimated magnitude of the historic increase in photosynthesis as result of increasing atmospheric CO2 concentrations, however, differs by an order of magnitude between long-term proxies and terrestrial biosphere models7-13. Here we quantify the historic effect of CO2 on global photosynthesis by identifying an emergent constraint14-16 that combines terrestrial biosphere models with global carbon budget estimates. Our analysis suggests that CO2 fertilization increased global annual photosynthesis by 11.85 ± 1.4%, or 13.98 ± 1.63 petagrams carbon (mean ± 95% confidence interval) between 1981 and 2020. Our results help resolve conflicting estimates of the historic sensitivity of global photosynthesis to CO2, and highlight the large impact anthropogenic emissions have had on ecosystems worldwide.

A bimodal burst energy distribution of a repeating fast radio burst source.

The event rate, energy distribution and time-domain behaviour of repeating fast radio bursts (FRBs) contain essential information regarding their physical nature and central engine, which are as yet unknown1,2. As the first precisely localized source, FRB 121102 (refs. 3-5) has been extensively observed and shows non-Poisson clustering of bursts over time and a power-law energy distribution6-8. However, the extent of the energy distribution towards the fainter end was not known. Here we report the detection of 1,652 independent bursts with a peak burst rate of 122 h-1, in 59.5 hours spanning 47 days. A peak in the isotropic equivalent energy distribution is found to be approximately 4.8 × 1037 erg at 1.25 GHz, below which the detection of bursts is suppressed. The burst energy distribution is bimodal, and well characterized by a combination of a log-normal function and a generalized Cauchy function. The large number of bursts in hour-long spans allows sensitive periodicity searches between 1 ms and 1,000 s. The non-detection of any periodicity or quasi-periodicity poses challenges for models involving a single rotating compact object. The high burst rate also implies that FRBs must be generated with a high radiative efficiency, disfavouring emission mechanisms with large energy requirements or contrived triggering conditions.

Quantum circuits with many photons on a programmable nanophotonic chip.

Growing interest in quantum computing for practical applications has led to a surge in the availability of programmable machines for executing quantum algorithms1,2. Present-day photonic quantum computers3-7 have been limited either to non-deterministic operation, low photon numbers and rates, or fixed random gate sequences. Here we introduce a full-stack hardware-software system for executing many-photon quantum circuit operations using integrated nanophotonics: a programmable chip, operating at room temperature and interfaced with a fully automated control system. The system enables remote users to execute quantum algorithms that require up to eight modes of strongly squeezed vacuum initialized as two-mode squeezed states in single temporal modes, a fully general and programmable four-mode interferometer, and photon number-resolving readout on all outputs. Detection of multi-photon events with photon numbers and rates exceeding any previous programmable quantum optical demonstration is made possible by strong squeezing and high sampling rates. We verify the non-classicality of the device output, and use the platform to carry out proof-of-principle demonstrations of three quantum algorithms: Gaussian boson sampling, molecular vibronic spectra and graph similarity8. These demonstrations validate the platform as a launchpad for scaling photonic technologies for quantum information processing.

Collaborators or competitors: the communication between RNA polymerase II and the nucleosome during eukaryotic transcription.

Decades of scientific research have been devoted to unraveling the intricacies of eukaryotic transcription since the groundbreaking discovery of eukaryotic RNA polymerases in the late 1960s. RNA polymerase II, the polymerase responsible for mRNA synthesis, has always attracted the most attention. Despite its structural resemblance to its bacterial counterpart, eukaryotic RNA polymerase II faces a unique challenge in progressing transcription due to the presence of nucleosomes that package DNA in the nuclei. In this review, we delve into the impact of RNA polymerase II and histone signaling on the progression of eukaryotic transcription. We explore the pivotal points of interactions that bridge the RNA polymerase II and histone signaling systems. Finally, we present an analysis of recent cryo-electron microscopy structures, which captured RNA polymerase II-nucleosome complexes at different stages of the transcription cycle. The combination of the signaling crosstalk and the direct visualization of RNA polymerase II-nucleosome complexes provides a deeper understanding of the communication between these two major players in eukaryotic transcription.

Electrical switching of magnetic order in an orbital Chern insulator.

Magnetism typically arises from the joint effect of Fermi statistics and repulsive Coulomb interactions, which favours ground states with non-zero electron spin. As a result, controlling spin magnetism with electric fields-a longstanding technological goal in spintronics and multiferroics1,2-can be achieved only indirectly. Here we experimentally demonstrate direct electric-field control of magnetic states in an orbital Chern insulator3-6, a magnetic system in which non-trivial band topology favours long-range order of orbital angular momentum but the spins are thought to remain disordered7-14. We use van der Waals heterostructures consisting of a graphene monolayer rotationally faulted with respect to a Bernal-stacked bilayer to realize narrow and topologically non-trivial valley-projected moiré minibands15-17. At fillings of one and three electrons per moiré unit cell within these bands, we observe quantized anomalous Hall effects18 with transverse resistance approximately equal to h/2e2 (where h is Planck's constant and e is the charge on the electron), which is indicative of spontaneous polarization of the system into a single-valley-projected band with a Chern number equal to two. At a filling of three electrons per moiré unit cell, we find that the sign of the quantum anomalous Hall effect can be reversed via field-effect control of the chemical potential; moreover, this transition is hysteretic, which we use to demonstrate non-volatile electric-field-induced reversal of the magnetic state. A theoretical analysis19 indicates that the effect arises from the topological edge states, which drive a change in sign of the magnetization and thus a reversal in the favoured magnetic state. Voltage control of magnetic states can be used to electrically pattern non-volatile magnetic-domain structures hosting chiral edge states, with applications ranging from reconfigurable microwave circuit elements to ultralow-power magnetic memories.

Enhanced intrinsic photovoltaic effect in tungsten disulfide nanotubes.

The photovoltaic effect in traditional p-n junctions-where a p-type material (with an excess of holes) abuts an n-type material (with an excess of electrons)-involves the light-induced creation of electron-hole pairs and their subsequent separation, generating a current. This photovoltaic effect is particularly important for environmentally benign energy harvesting, and its efficiency has been increased dramatically, almost reaching the theoretical limit1. Further progress is anticipated by making use of the bulk photovoltaic effect (BPVE)2, which does not require a junction and occurs only in crystals with broken inversion symmetry3. However, the practical implementation of the BPVE is hampered by its low efficiency in existing materials4-10. Semiconductors with reduced dimensionality2 or a smaller bandgap4,5 have been suggested to be more efficient. Transition-metal dichalcogenides (TMDs) are exemplary small-bandgap, two-dimensional semiconductors11,12 in which various effects have been observed by breaking the inversion symmetry inherent in their bulk crystals13-15, but the BPVE has not been investigated. Here we report the discovery of the BPVE in devices based on tungsten disulfide, a member of the TMD family. We find that systematically reducing the crystal symmetry beyond mere broken inversion symmetry-moving from a two-dimensional monolayer to a nanotube with polar properties-greatly enhances the BPVE. The photocurrent density thus generated is orders of magnitude larger than that of other BPVE materials. Our findings highlight not only the potential of TMD-based nanomaterials, but also more generally the importance of crystal symmetry reduction in enhancing the efficiency of converting solar to electric power.

Targeting of REST with rationally-designed small molecule compounds exhibits synergetic therapeutic potential in human glioblastoma cells.

BACKGROUND: Glioblastoma (GBM) is an aggressive brain cancer associated with poor prognosis, intrinsic heterogeneity, plasticity, and therapy resistance. In some GBMs, cell proliferation is fueled by a transcriptional regulator, repressor element-1 silencing transcription factor (REST). RESULTS: Using CRISPR/Cas9, we identified GBM cell lines dependent on REST activity. We developed new small molecule inhibitory compounds targeting small C-terminal domain phosphatase 1 (SCP1) to reduce REST protein level and transcriptional activity in glioblastoma cells. Top leads of the series like GR-28 exhibit potent cytotoxicity, reduce REST protein level, and suppress its transcriptional activity. Upon the loss of REST protein, GBM cells can potentially compensate by rewiring fatty acid metabolism, enabling continued proliferation. Combining REST inhibition with the blockade of this compensatory adaptation using long-chain acyl-CoA synthetase inhibitor Triacsin C demonstrated substantial synergetic potential without inducing hepatotoxicity. CONCLUSIONS: Our results highlight the efficacy and selectivity of targeting REST alone or in combination as a therapeutic strategy to combat high-REST GBM.

Enhancing the Humidity Resistance of Triboelectric Nanogenerators: A Review.

Triboelectric nanogenerators (TENGs) are sustainable energy resources for powering electronic devices from miniature to large-scale applications. However, their output performance and stability can deteriorate significantly when TENGs are exposed to moisture or humidity caused by the ambient environment or human physiological activities. This review provides an overview of the recent research advancements in enhancing the humidity resistance of TENGs. Various approaches have been reviewed including encapsulation techniques, surface modification of triboelectric materials to augment hydrophobicity or superhydrophobicity, the creation of fibrous architectures for effective moisture dissipation, leveraging water assistance for TENG performance enhancement, and other strategies like charge excitation. These research efforts contribute to the improvement of environmental adaptability and lead to expanded practical TENG applications both as energy harvesters and self-powered sensors. The efficacy of these strategies and future challenges are also discussed to facilitate the continued development of resilient TENGs in high humidity environments.

Photothermal-Controllable Microneedles with Antitumor, Antioxidant, Angiogenic, and Chondrogenic Activities to Sequential Eliminate Tracheal Neoplasm and Reconstruct Tracheal Cartilage.

The optimal treatment for tracheal tumors necessitates sequential tumor elimination and tracheal cartilage reconstruction. This study introduces an innovative inorganic nanosheet, MnO2 /PDA@Cu, comprising manganese dioxide (MnO2 ) loaded with copper ions (Cu) through in situ polymerization using polydopamine (PDA) as an intermediary. Additionally, a specialized methacrylic anhydride modified decellularized cartilage matrix (MDC) hydrogel with chondrogenic effects is developed by modifying a decellularized cartilage matrix with methacrylic anhydride. The MnO2 /PDA@Cu nanosheet is encapsulated within MDC-derived microneedles, creating a photothermal-controllable MnO2 /PDA@Cu-MDC microneedle. Effectiveness evaluation involved deep insertion of the MnO2 /PDA@Cu-MDC microneedle into tracheal orthotopic tumor in a murine model. Under 808 nm near-infrared irradiation, facilitated by PDA, the microneedle exhibited rapid overheating, efficiently eliminating tumors. PDA's photothermal effects triggered controlled MnO2 and Cu release. The MnO2 nanosheet acted as a potent inorganic nanoenzyme, scavenging reactive oxygen species for an antioxidant effect, while Cu facilitated angiogenesis. This intervention enhanced blood supply at the tumor excision site, promoting stem cell enrichment and nutrient provision. The MDC hydrogel played a pivotal role in creating a chondrogenic niche, fostering stem cells to secrete cartilaginous matrix. In conclusion, the MnO2 /PDA@Cu-MDC microneedle is a versatile platform with photothermal control, sequentially combining antitumor, antioxidant, pro-angiogenic, and chondrogenic activities to orchestrate precise tracheal tumor eradication and cartilage regeneration.

ForMAX - a beamline for multiscale and multimodal structural characterization of hierarchical materials.

The ForMAX beamline at the MAX IV Laboratory provides multiscale and multimodal structural characterization of hierarchical materials in the nanometre to millimetre range by combining small- and wide-angle X-ray scattering with full-field microtomography. The modular design of the beamline is optimized for easy switching between different experimental modalities. The beamline has a special focus on the development of novel fibrous materials from forest resources, but it is also well suited for studies within, for example, food science and biomedical research.

Combining single-cell tracking and omics improves blood stem cell fate regulator identification.

Molecular programs initiating cell fate divergence (CFD) are difficult to identify. Current approaches usually compare cells long after CFD initiation, therefore missing molecular changes at its start. Ideally, single cells that differ in their CFD molecular program but are otherwise identical are compared early in CFD. This is possible in diverging sister cells, which were identical until their mother's division and thus differ mainly in CFD properties. In asymmetrically dividing cells, divergent daughter fates are prospectively committed during division, and diverging sisters can thus be identified at the start of CFD. Using asymmetrically dividing blood stem cells, we developed a pipeline (ie, trackSeq) for imaging, tracking, isolating, and transcriptome sequencing of single cells. Their identities, kinship, and histories are maintained throughout, massively improving molecular noise filtering and candidate identification. In addition to many identified blood stem CFD regulators, we offer here this pipeline for use in CFDs other than asymmetric division.

New radiofrequency ablation procedure for selective reduction in complicated monochorionic twin or triplet pregnancy using multistep, incremental expansion technique.

OBJECTIVE: Radiofrequency ablation (RFA) is the preferred approach for selective reduction in complex monochorionic (MC) multiple pregnancies owing to the ease of operation and minimal invasiveness. To optimize the RFA technique and reduce the risk of adverse pregnancy outcome resulting from the heat-sink effect of RFA therapy, we used an innovative RFA method, in which an electrode needle was expanded incrementally and stepwise. This study aimed to assess the efficacy and safety profile of this novel multistep incremental expansion RFA method for selective fetal reduction in MC twin and triplet pregnancies. METHODS: This was a single-center retrospective cohort study of all MC multiple pregnancies undergoing RFA between March 2016 and October 2022 at our center. The multistep RFA technique involved the use of an expandable needle, which was gradually expanded during the RFA procedure until cessation of umbilical cord blood flow was achieved. The needle used for the single-step RFA method was fully extended from the start of treatment. RESULTS: In total, 132 MC multiple pregnancies underwent selective reduction using RFA, including 50 cases undergoing multistep RFA and 82 cases undergoing single-step RFA. The overall survival rates were not significantly different between the multistep and single-step RFA groups (81.1% vs 72.3%; P = 0.234). Similarly, the rates of preterm prelabor rupture of the membranes within 2 weeks after RFA, procedure-related complications, spontaneous preterm delivery and pathological findings on cranial ultrasound, as well as gestational age at delivery and birth weight, did not differ between the two groups. However, there was a trend towards a prolonged procedure-to-delivery interval following multistep RFA compared with single-step RFA (median, 109 vs 99 days; P = 0.377). Moreover, the fetal loss rate within 2 weeks after RFA in the multistep RFA group was significantly lower than that in the single-step RFA group (10.0% vs 24.4%; P = 0.041). The median ablation time was shorter (5.3 vs 7.8 min; P < 0.001) and the median ablation energy was lower (10.2 vs 18.0 kJ; P < 0.001) in multistep compared with single-step RFA. There were no significant differences in neonatal outcomes following multistep vs single-step RFA. CONCLUSIONS: Overall survival rates were similar between the two RFA methods. However, the multistep RFA technique was associated with a lower risk of fetal loss within 2 weeks after RFA. The multistep RFA technique required significantly less ablation energy and a shorter ablation time compared with single-step RFA in selective fetal reduction of MC twin and triplet pregnancies. Additionally, there was a trend towards a prolonged procedure-to-delivery interval with the multistep RFA technique. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Value of micro-flow imaging and high-definition micro-flow imaging in differentiating malignant and benign breast lesions.

AIM: To evaluate the value of non-contrast micro-flow imaging (MFI) and high-definition micro-flow imaging (HD-MFI) in differentiating malignant and benign breast lesions. MATERIALS AND METHODS: One hundred and thirty-three patients with 138 breast lesions (80 benign and 58 malignant lesions) were examined using colour Doppler flow imaging (CDFI), MFI, and HD-MFI before biopsy, with blood flow signals graded into four types (grade 0, 1, 2, and 3) and penetrating vessels evaluated. The micro-vascular patterns of MFI and HD-MFI were evaluated and classified into five patterns: avascular, line-like, tree-like, root hair-like, and crab claw-like pattern. The diagnostic efficiency of micro-vascular patterns was analysed. Moreover, ultrasound Breast Imaging Reporting and Data System (BI-RADS) 4A lesions were also re-assessed according to the micro-vascular patterns of MFI or HD-MFI. RESULTS: The capability of detecting blood flow and penetrating vessels from high to low was HD-MFI, MFI, and CDFI, respectively (p<0.05). Rich blood flow signals, penetrating vessels, and root hair-like or crab claw-like pattern were more likely in malignant breast lesions, while few blood flow signals, tree-like pattern were mostly in benign lesions (p<0.05). The diagnostic efficiency of HD-MFI and MFI were higher than CDFI (p>0.05). MFI could reduce unnecessary biopsy of 52 US BI-RADS 4A lesions but with two malignancies missed, while 56 ultrasound BI-RADS 4A lesions could be downgraded by HD-MFI with none malignancies missed. CONCLUSIONS: MFI and HD-MFI can detect more blood flow in breast lesions than CDFI, and could help distinguish benign and malignant breast lesions. HD-MFI could reduce the unnecessary biopsy of US BI-RADS 4A lesions without missed malignancy.

Sub-region based radiomics analysis for prediction of isocitrate dehydrogenase and telomerase reverse transcriptase promoter mutations in diffuse gliomas.

AIM: To enhance the prediction of mutation status of isocitrate dehydrogenase (IDH) and telomerase reverse transcriptase (TERT) promoter, which are crucial for glioma prognostication and therapeutic decision-making, via sub-regional radiomics analysis based on multiparametric magnetic resonance imaging (MRI). MATERIALS AND METHODS: A retrospective study was conducted on 401 participants with adult-type diffuse gliomas. Employing the K-means algorithm, tumours were clustered into two to four subregions. Sub-regional radiomics features were extracted and selected using the Mann-Whitney U-test, Pearson correlation analysis, and least absolute shrinkage and selection operator, forming the basis for predictive models. The performance of model combinations of different sub-regional features and classifiers (including logistic regression, support vector machines, K-nearest neighbour, light gradient boosting machine, and multilayer perceptron) was evaluated using an external test set. RESULTS: The models demonstrated high predictive performance, with area under the receiver operating characteristic curve (AUC) values ranging from 0.918 to 0.994 in the training set for IDH mutation prediction and from 0.758 to 0.939 for TERT promoter mutation prediction. In the external test sets, the two-cluster radiomics features and the logistic regression model yielded the highest prediction for IDH mutation, resulting in an AUC of 0.905. Additionally, the most effective predictive performance with an AUC of 0.803 was achieved using the four-cluster radiomics features and the support vector machine model, specifically for TERT promoter mutation prediction. CONCLUSION: The present study underscores the potential of sub-regional radiomics analysis in predicting IDH and TERT promoter mutations in glioma patients. These models have the capacity to refine preoperative glioma diagnosis and contribute to personalised therapeutic interventions for patients.

"Very early" intrahepatic cholangiocarcinoma (≤ 2.0 cm): MRI manifestation and prognostic potential.

AIMS: To explore the MRI characteristics and clinical outcome of the "very early" intrahepatic cholangiocarcinoma (iCCA) ≤2.0cm. MATERIALS AND METHODS: Totally 213 pathologically confirmed iCCAs (44 ≤ 2.0cm and 169 of 2.0-5.0cm) from two institutes were included. Forty-four matching non-iCCA malignancies ≤2.0cm were also enrolled. Recurrence-free survival (RFS) was estimated and compared between iCCAs ≤2.0cm and 2.0-5.0cm. MRI features were analyzed and compared between iCCAs ≤2.0cm and 2.0-5.0cm, as well as between iCCAs ≤2.0cm and non-iCCAs ≤2.0cm. Univariate and multivariate regression analyses were performed to identify independent imaging features for discrimination. An MRI-based diagnostic model for iCCA ≤2.0cm was constructed by incorporating the independent imaging features. RESULTS: ICCAs ≤2.0cm had a significantly longer RFS than those of 2.0-5.0cm (log rank P=0.014). Imaging features of homogeneous signal (odds ratio (OR) = 6.677, P<0.001) and lack of vessel invasion (OR=7.56, P<0.001) were more frequently displayed in iCCAs ≤2.0cm compared to iCCAs of 2.0-5.0cm independently. In the small lesions ≤2.0cm, imaging features of progressive or persistent enhancement pattern (OR=27.78, P=0.002) and rim diffusion restriction (OR=5.70, P=0.027) were independent imaging features suggestive of iCCA over non-iCCA malignancy; their combination yielded an area under the curve value of 0.824, with a sensitivity of 97.73%. CONCLUSION: The "very early" iCCA ≤2.0cm was associated with a favorable outcome after surgery, it displayed different and relatively atypical imaging manifestations compared with those of 2.0-5.0cm. Furthermore, in the small lesions ≤ 2.0cm, MRI can be served as a useful non-invasive diagnostic tool for iCCA in clinical screening with high sensitivity.

Inter-equipment validation of PET-based radiomics for predicting EGFR mutation statuses in patients with non-small cell lung cancer.

AIM: To validate the inter-equipment generality of the radiomics based on PET images to predict the EGFR mutation status of patients with non-small cell lung cancer. MATERIALS AND METHODS: Patients were retrospectively collected in the departments of nuclear medicine of Heyi branch (Siemens equipment) and East branch (General Electric (GE) equipment) of the first affiliated hospital of Zhengzhou university. 5 predicting logistic regression models were established. The 1st one was trained and tested by the GE dataset; The 2nd one was trained and tested by the Siemens dataset; The 3rd one was trained and tested by the mixed dataset consisting of GE and Siemens. The 4th one was trained by GE and tested by Siemens; The 5th one was trained by Siemens and tested by GE. RESULTS: For the 1st ∼ 5th models, the mean values of AUCs for training/testing datasets were 0.78/0.73, 0.74/0.72, 0.75/0.70, 0.74/0.65 and 0.68/0.63, respectively. CONCLUSION: The AUCs of the models trained and tested on the datasets from the same equipment were higher than those for different equipment. The inter-equipment generality of the radiomics was not good enough in clinical practice.

Deep-learning features based on F18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) to predict preoperative colorectal cancer lymph node metastasis.

AIM: The objective of this study was to create and authenticate a prognostic model for lymph node metastasis (LNM) in colorectal cancer (CRC) that integrates clinical, radiomics, and deep transfer learning features. MATERIALS AND METHODS: In this study, we analyzed data from 119 CRC patients who underwent F18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) scanning. The patient cohort was divided into training and validation subsets in an 8:2 ratio, with an additional 33 external data points for testing. Initially, we conducted univariate analysis to screen clinical parameters. Radiomics features were extracted from manually drawn images using pyradiomics, and deep-learning features, radiomics features, and clinical features were selected using Least Absolute Shrinkage and Selection Operator (LASSO) regression and Spearman correlation coefficient. We then constructed a model by training a support vector machine (SVM), and evaluated the performance of the prediction model by comparing the area under the curve (AUC), sensitivity, and specificity. Finally, we developed nomograms combining clinical and radiological features for interpretation and analysis. RESULTS: The deep learning radiomics (DLR) nomogram model, which was developed by integrating deep learning, radiomics, and clinical features, exhibited excellent performance. The area under the curve was (AUC = 0.934, 95% confidence interval [CI]: 0.884-0.983) in the training cohort, (AUC = 0.902, 95% CI: 0.769-1.000) in the validation cohort, and (AUC = 0.836, 95% CI: 0.673-0.998) in the test cohort. CONCLUSION: We developed a preoperative predictive machine-learning model using deep transfer learning, radiomics, and clinical features to differentiate LNM status in CRC, aiding in treatment decision-making for patients.