Improving Pancreatic Cyst Management: Artificial Intelligence-Powered Prediction of Advanced Neoplasms through Endoscopic Ultrasound-Guided Confocal Endomicroscopy.

Despite the increasing rate of detection of incidental pancreatic cystic lesions (PCLs), current standard-of-care methods for their diagnosis and risk stratification remain inadequate. Intraductal papillary mucinous neoplasms (IPMNs) are the most prevalent PCLs. The existing modalities, including endoscopic ultrasound and cyst fluid analysis, only achieve accuracy rates of 65-75% in identifying carcinoma or high-grade dysplasia in IPMNs. Furthermore, surgical resection of PCLs reveals that up to half exhibit only low-grade dysplastic changes or benign neoplasms. To reduce unnecessary and high-risk pancreatic surgeries, more precise diagnostic techniques are necessary. A promising approach involves integrating existing data, such as clinical features, cyst morphology, and data from cyst fluid analysis, with confocal endomicroscopy and radiomics to enhance the prediction of advanced neoplasms in PCLs. Artificial intelligence and machine learning modalities can play a crucial role in achieving this goal. In this review, we explore current and future techniques to leverage these advanced technologies to improve diagnostic accuracy in the context of PCLs.

Artificial Intelligence in the Diagnosis and Treatment of Pancreatic Cystic Lesions and Adenocarcinoma.

Pancreatic cancer is projected to become the second leading cause of cancer-related mortality in the United States by 2030. This is in part due to the paucity of reliable screening and diagnostic options for early detection. Amongst known pre-malignant pancreatic lesions, pancreatic intraepithelial neoplasia (PanIN) and intraductal papillary mucinous neoplasms (IPMNs) are the most prevalent. The current standard of care for the diagnosis and classification of pancreatic cystic lesions (PCLs) involves cross-sectional imaging studies and endoscopic ultrasound (EUS) and, when indicated, EUS-guided fine needle aspiration and cyst fluid analysis. However, this is suboptimal for the identification and risk stratification of PCLs, with accuracy of only 65-75% for detecting mucinous PCLs. Artificial intelligence (AI) is a promising tool that has been applied to improve accuracy in screening for solid tumors, including breast, lung, cervical, and colon cancer. More recently, it has shown promise in diagnosing pancreatic cancer by identifying high-risk populations, risk-stratifying premalignant lesions, and predicting the progression of IPMNs to adenocarcinoma. This review summarizes the available literature on artificial intelligence in the screening and prognostication of precancerous lesions in the pancreas, and streamlining the diagnosis of pancreatic cancer.

Active and Low-Cost Hyperspectral Imaging for the Spectral Analysis of a Low-Light Environment.

Hyperspectral imaging is capable of capturing information beyond conventional RGB cameras; therefore, several applications of this have been found, such as material identification and spectral analysis. However, similar to many camera systems, most of the existing hyperspectral cameras are still passive imaging systems. Such systems require an external light source to illuminate the objects, to capture the spectral intensity. As a result, the collected images highly depend on the environment lighting and the imaging system cannot function in a dark or low-light environment. This work develops a prototype system for active hyperspectral imaging, which actively emits diverse single-wavelength light rays at a specific frequency when imaging. This concept has several advantages: first, using the controlled lighting, the magnitude of the individual bands is more standardized to extract reflectance information; second, the system is capable of focusing on the desired spectral range by adjusting the number and type of LEDs; third, an active system could be mechanically easier to manufacture, since it does not require complex band filters as used in passive systems. Three lab experiments show that such a design is feasible and could yield informative hyperspectral images in low light or dark environments: (1) spectral analysis: this system's hyperspectral images improve food ripening and stone type discernibility over RGB images; (2) interpretability: this system's hyperspectral images improve machine learning accuracy. Therefore, it can potentially benefit the academic and industry segments, such as geochemistry, earth science, subsurface energy, and mining.

Two-Stage Mesh Deep Learning for Automated Tooth Segmentation and Landmark Localization on 3D Intraoral Scans.

Accurately segmenting teeth and identifying the corresponding anatomical landmarks on dental mesh models are essential in computer-aided orthodontic treatment. Manually performing these two tasks is time-consuming, tedious, and, more importantly, highly dependent on orthodontists' experiences due to the abnormality and large-scale variance of patients' teeth. Some machine learning-based methods have been designed and applied in the orthodontic field to automatically segment dental meshes (e.g., intraoral scans). In contrast, the number of studies on tooth landmark localization is still limited. This paper proposes a two-stage framework based on mesh deep learning (called TS-MDL) for joint tooth labeling and landmark identification on raw intraoral scans. Our TS-MDL first adopts an end-to-end iMeshSegNet method (i.e., a variant of the existing MeshSegNet with both improved accuracy and efficiency) to label each tooth on the downsampled scan. Guided by the segmentation outputs, our TS-MDL further selects each tooth's region of interest (ROI) on the original mesh to construct a light-weight variant of the pioneering PointNet (i.e., PointNet-Reg) for regressing the corresponding landmark heatmaps. Our TS-MDL was evaluated on a real-clinical dataset, showing promising segmentation and localization performance. Specifically, iMeshSegNet in the first stage of TS-MDL reached an averaged Dice similarity coefficient (DSC) at 0.964±0.054 , significantly outperforming the original MeshSegNet. In the second stage, PointNet-Reg achieved a mean absolute error (MAE) of 0.597±0.761 mm in distances between the prediction and ground truth for 66 landmarks, which is superior compared with other networks for landmark detection. All these results suggest the potential usage of our TS-MDL in orthodontics.

Application of Artificial Intelligence in the Management of Pancreatic Cystic Lesions.

The rate of incidentally detected pancreatic cystic lesions (PCLs) has increased over the past decade and was recently reported at 8%. These lesions pose a unique challenge, as each subtype of PCL carries a different risk of malignant transformation, ranging from 0% (pancreatic pseudocyst) to 34-68% (main duct intraductal papillary mucinous neoplasm). It is imperative to correctly risk-stratify the malignant potential of these lesions in order to provide the correct care course for the patient, ranging from monitoring to surgical intervention. Even with the multiplicity of guidelines (i.e., the American Gastroenterology Association guidelines and Fukuoka/International Consensus guidelines) and multitude of diagnostic information, risk stratification of PCLs falls short. Studies have reported that 25-64% of patients undergoing PCL resection have pancreatic cysts with no malignant potential, and up to 78% of mucin-producing cysts resected harbor no malignant potential on pathological evaluation. Clinicians are now incorporating artificial intelligence technology to aid in the management of these difficult lesions. This review article focuses on advancements in artificial intelligence within digital pathomics, radiomics, and genomics as they apply to the diagnosis and risk stratification of PCLs.

Few-Shot Learning with a Strong Teacher.

Few-shot learning (FSL) aims to generate a classifier using limited labeled examples. Many existing works take the meta-learning approach, constructing a few-shot learner (a meta-model) that can learn from few-shot examples to generate a classifier. The performance is measured by how well the resulting classifiers classify the test (\ie, query) examples of those tasks. In this paper, we point out two potential weaknesses of this approach. First, the sampled query examples may not provide sufficient supervision for meta-training the few-shot learner. Second, the effectiveness of meta-learning diminishes sharply with the increasing number of shots. We propose a novel meta-training objective for the few-shot learner, which is to encourage the few-shot learner to generate classifiers that perform like strong classifiers. Concretely, we associate each sampled few-shot task with a strong classifier, which is trained with ample labeled examples. The strong classifiers can be seen as the target classifiers that we hope the few-shot learner to generate given few-shot examples, and we use the strong classifiers to supervise the few-shot learner. We validate our approach in combinations with many representative meta-learning methods. More importantly, with our approach, meta-learning based FSL methods can consistently outperform non-meta-learning based methods at different numbers of shots.

A Comparison of Machine Learning Methods and Conventional Logistic Regression for the Prediction of In-Hospital Mortality in Acute Biliary Pancreatitis.

OBJECTIVES: For population databases, multivariable regressions are established analytical standards. The utilization of machine learning (ML) in population databases is novel. We compared conventional statistical methods and ML for predicting mortality in biliary acute pancreatitis (biliary AP). METHODS: Using the Nationwide Readmission Database (2010-2014), we identified patients (age ≥18 years) with admissions for biliary AP. These data were randomly divided into a training (70%) and test set (30%), stratified by the outcome of mortality. The accuracy of ML and logistic regression models in predicting mortality was compared using 3 different assessments. RESULTS: Among 97,027 hospitalizations for biliary AP, mortality rate was 0.97% (n = 944). Predictors of mortality included severe AP, sepsis, increasing age, and nonperformance of cholecystectomy. Assessment metrics for predicting the outcome of mortality, the scaled Brier score (odds ratio [OR], 0.24; 95% confidence interval [CI], 0.16-0.33 vs 0.18; 95% CI, 0.09-0.27), F-measure (OR, 43.4; 95% CI, 38.3-48.6 vs 40.6; 95% CI, 35.7-45.5), and the area under the receiver operating characteristic (OR, 0.96; 95% CI, 0.94-0.97 vs 0.95; 95% CI, 0.94-0.96) were comparable between the ML and logistic regression models, respectively. CONCLUSIONS: For population databases, traditional multivariable analysis is noninferior to ML-based algorithms in predictive modeling of hospital outcomes for biliary AP.

A General and Predictive Understanding of Thermal Transport from 1D- and 2D-Confined Nanostructures: Theory and Experiment.

Heat management is crucial in the design of nanoscale devices as the operating temperature determines their efficiency and lifetime. Past experimental and theoretical works exploring nanoscale heat transport in semiconductors addressed known deviations from Fourier's law modeling by including effective parameters, such as a size-dependent thermal conductivity. However, recent experiments have qualitatively shown behavior that cannot be modeled in this way. Here, we combine advanced experiment and theory to show that the cooling of 1D- and 2D-confined nanoscale hot spots on silicon can be described using a general hydrodynamic heat transport model, contrary to previous understanding of heat flow in bulk silicon. We use a comprehensive set of extreme ultraviolet scatterometry measurements of nondiffusive transport from transiently heated nanolines and nanodots to validate and generalize our ab initio model, that does not need any geometry-dependent fitting parameters. This allows us to uncover the existence of two distinct time scales and heat transport mechanisms: an interface resistance regime that dominates on short time scales and a hydrodynamic-like phonon transport regime that dominates on longer time scales. Moreover, our model can predict the full thermomechanical response on nanometer length scales and picosecond time scales for arbitrary geometries, providing an advanced practical tool for thermal management of nanoscale technologies. Furthermore, we derive analytical expressions for the transport time scales, valid for a subset of geometries, supplying a route for optimizing heat dissipation.

Machine learning from clinical data sets of a contemporary decision for orthodontic tooth extraction.

OBJECTIVE: To examine the robustness of the published machine learning models in the prediction of extraction vs non-extraction for a diverse US sample population seen by multiple providers. SETTING AND SAMPLE POPULATION: Diverse group of 838 patients (208 extraction, 630 non-extraction) were consecutively enrolled. MATERIALS AND METHODS: Two sets of input features (117 and 22) including clinical and cephalometric variables were identified based on previous studies. Random forest (RF) and multilayer perception (MLP) models were trained using these feature sets on the sample population and evaluated using measures including accuracy (ACC) and balanced accuracy (BA). A technique to identify incongruent data was used to explore underlying characteristics of the data set and split all samples into 2 groups (G1 and G2) for further model training. RESULTS: Performance of the models (75%-79% ACC and 72%-76% BA) on the total sample population was lower than in previous research. Models were retrained and evaluated using G1 and G2 separately, and individual group MLP models yielded improved accuracy for G1 (96% ACC and 94% BA) and G2 (88% ACC and 85% BA). RF feature ranking showed differences between top features for G1 (maxillary crowding, mandibular crowding and L1-NB) and G2 (age, mandibular crowding and lower lip to E-plane). CONCLUSIONS: An incongruent data pattern exists in a consecutively enrolled patient population. Future work with incongruent data segregation and advanced artificial intelligence algorithms is needed to improve the generalization ability to make it ready to support clinical decision-making.

Targeted Writing and Deleting of Magnetic Skyrmions in Two-Terminal Nanowire Devices.

Controllable writing and deleting of nanoscale magnetic skyrmions are key requirements for their use as information carriers for next-generation memory and computing technologies. While several schemes have been proposed, they require complex fabrication techniques or precisely tailored electrical inputs, which limits their long-term scalability. Here, we demonstrate an alternative approach for writing and deleting skyrmions using conventional electrical pulses within a simple, two-terminal wire geometry. X-ray microscopy experiments and micromagnetic simulations establish the observed skyrmion creation and annihilation as arising from Joule heating and Oersted field effects of the current pulses, respectively. The unique characteristics of these writing and deleting schemes, such as spatial and temporal selectivity, together with the simplicity of the two-terminal device architecture, provide a flexible and scalable route to the viable applications of skyrmions.

High performance in risk stratification of intraductal papillary mucinous neoplasms by confocal laser endomicroscopy image analysis with convolutional neural networks (with video).

BACKGROUND AND AIMS: EUS-guided needle-based confocal laser endomicroscopy (EUS-nCLE) can differentiate high-grade dysplasia/adenocarcinoma (HGD-Ca) in intraductal papillary mucinous neoplasms (IPMNs) but requires manual interpretation. We sought to derive predictive computer-aided diagnosis (CAD) and artificial intelligence (AI) algorithms to facilitate accurate diagnosis and risk stratification of IPMNs. METHODS: A post hoc analysis of a single-center prospective study evaluating EUS-nCLE (2015-2019; INDEX study) was conducted using 15,027 video frames from 35 consecutive patients with histopathologically proven IPMNs (18 with HGD-Ca). We designed 2 CAD-convolutional neural network (CNN) algorithms: (1) a guided segmentation-based model (SBM), where the CNN-AI system was trained to detect and measure papillary epithelial thickness and darkness (indicative of cellular and nuclear stratification), and (2) a reasonably agnostic holistic-based model (HBM) where the CNN-AI system automatically extracted nCLE features for risk stratification. For the detection of HGD-Ca in IPMNs, the diagnostic performance of the CNN-CAD algorithms was compared with that of the American Gastroenterological Association (AGA) and revised Fukuoka guidelines. RESULTS: Compared with the guidelines, both n-CLE-guided CNN-CAD algorithms yielded higher sensitivity (HBM, 83.3%; SBM, 83.3%; AGA, 55.6%; Fukuoka, 55.6%) and accuracy (SBM, 82.9%; HBM, 85.7%; AGA, 68.6%; Fukuoka, 74.3%) for diagnosing HGD-Ca, with comparable specificity (SBM, 82.4%; HBM, 88.2%; AGA, 82.4%; Fukuoka, 94.1%). Both CNN-CAD algorithms, the guided (SBM) and agnostic (HBM) models, were comparable in risk stratifying IPMNs. CONCLUSION: EUS-nCLE-based CNN-CAD algorithms can accurately risk stratify IPMNs. Future multicenter validation studies and AI model improvements could enhance the accuracy and fully automatize the process for real-time interpretation.

An ultrahigh-resolution soft x-ray microscope for quantitative analysis of chemically heterogeneous nanomaterials.

The analysis of chemical states and morphology in nanomaterials is central to many areas of science. We address this need with an ultrahigh-resolution scanning transmission soft x-ray microscope. Our instrument provides multiple analysis tools in a compact assembly and can achieve few-nanometer spatial resolution and high chemical sensitivity via x-ray ptychography and conventional scanning microscopy. A novel scanning mechanism, coupled to advanced x-ray detectors, a high-brightness x-ray source, and high-performance computing for analysis provide a revolutionary step forward in terms of imaging speed and resolution. We present x-ray microscopy with 8-nm full-period spatial resolution and use this capability in conjunction with operando sample environments and cryogenic imaging, which are now routinely available. Our multimodal approach will find wide use across many fields of science and facilitate correlative analysis of materials with other types of probes.

Single-shot large field of view Fourier transform holography with a picosecond plasma-based soft X-ray laser.

It is challenging to obtain nanoscale resolution images in a single ultrafast shot because a large number of photons, greater than 1011, are required in a single pulse of the illuminating source. We demonstrate single-shot high resolution Fourier transform holography over a broad 7 µm diameter field of view with ∼ 5 ps temporal resolution. The experiment used a plasma-based soft X-ray laser operating at 18.9 nm wavelength with nearly full spatial coherence and close to diffraction-limited divergence implemented utilizing a dual-plasma amplifier scheme. A Fresnel zone plate with a central aperture is used to efficiently generate the object and reference beams. Rapid numerical reconstruction by a 2D Fourier transform allows for real-time imaging. A half-pitch spatial resolution of 62 nm was obtained. This single-shot nanoscale-resolution imaging technique will allow for real-time ultrafast imaging of dynamic phenomena in compact setups.

Direct Demonstration of Topological Stability of Magnetic Skyrmions via Topology Manipulation.

Topological protection precludes a continuous deformation between topologically inequivalent configurations in a continuum. Motivated by this concept, magnetic skyrmions, topologically nontrivial spin textures, are expected to exhibit topological stability, thereby offering a prospect as a nanometer-scale nonvolatile information carrier. In real materials, however, atomic spins are configured as not continuous but discrete distributions, which raises a fundamental question if the topological stability is indeed preserved for real magnetic skyrmions. Answering this question necessitates a direct comparison between topologically nontrivial and trivial spin textures, but the direct comparison in one sample under the same magnetic fields has been challenging. Here we report how to selectively achieve either a skyrmion state or a topologically trivial bubble state in a single specimen and thereby experimentally show how robust the skyrmion structure is in comparison with the bubbles. We demonstrate that topologically nontrivial magnetic skyrmions show longer lifetimes than trivial bubble structures, evidencing the topological stability in a real discrete system. Our work corroborates the physical importance of the topology in the magnetic materials, which has hitherto been suggested by mathematical arguments, providing an important step toward ever-dense and more-stable magnetic devices.

Switchable resolution in soft x-ray tomography of single cells.

The diversity of living cells, in both size and internal complexity, calls for imaging methods with adaptable spatial resolution. Soft x-ray tomography (SXT) is a three-dimensional imaging technique ideally suited to visualizing and quantifying the internal organization of single cells of varying sizes in a near-native state. The achievable resolution of the soft x-ray microscope is largely determined by the objective lens, but switching between objectives is extremely time-consuming and typically undertaken only during microscope maintenance procedures. Since the resolution of the optic is inversely proportional to the depth of focus, an optic capable of imaging the thickest cells is routinely selected. This unnecessarily limits the achievable resolution in smaller cells and eliminates the ability to obtain high-resolution images of regions of interest in larger cells. Here, we describe developments to overcome this shortfall and allow selection of microscope optics best suited to the specimen characteristics and data requirements. We demonstrate that switchable objective capability advances the flexibility of SXT to enable imaging cells ranging in size from bacteria to yeast and mammalian cells without physically modifying the microscope, and we demonstrate the use of this technology to image the same specimen with both optics.

Application of Artificial Intelligence in the Detection and Differentiation of Colon Polyps: A Technical Review for Physicians.

Research in computer-aided diagnosis (CAD) and the application of artificial intelligence (AI) in the endoscopic evaluation of the gastrointestinal tract is novel. Since colonoscopy and detection of polyps can decrease the risk of colon cancer, it is recommended by multiple national and international societies. However, the procedure of colonoscopy is performed by humans where there are significant interoperator and interpatient variations, and hence, the risk of missing detection of adenomatous polyps. Early studies involving CAD and AI for the detection and differentiation of polyps show great promise. In this appraisal, we review existing scientific aspects of AI in CAD of colon polyps and discuss the pitfalls and future directions for advancing the science. This review addresses the technical intricacies in a manner that physicians can comprehend to promote a better understanding of this novel application.

Dynamics of the Bloch point in an asymmetric permalloy disk.

A Bloch point (BP) is a topological defect in a ferromagnet at which the local magnetization vanishes. With the difficulty of generating a stable BP in magnetic nanostructures, the intrinsic nature of a BP and its dynamic behaviour has not been verified experimentally. We report a realization of steady-state BPs embedded in deformed magnetic vortex cores in asymmetrically shaped Ni80Fe20 nanodisks. Time-resolved nanoscale magnetic X-ray imaging combined with micromagnetic simulation shows detailed dynamic character of BPs, revealing rigid and limited lateral movements under magnetic field pulses as well as its crucial role in vortex-core dynamics. Direct visualizations of magnetic structures disclose the unique dynamical feature of a BP as an atomic scale discrete spin texture and allude its influence on the neighbouring spin structures such as magnetic vortices.

Soft x-ray spectroscopy of high pressure liquid.

We describe a new experimental technique that allows for soft x-ray spectroscopy studies (∼100-1000 eV) of high pressure liquid (∼100 bars). We achieve this through a liquid cell with a 100 nm-thick Si3N4 membrane window, which is sandwiched by two identical O-rings for vacuum sealing. The thin Si3N4 membrane allows soft x-rays to penetrate, while separating the high-pressure liquid under investigation from the vacuum required for soft x-ray transmission and detection. The burst pressure of the Si3N4 membrane increases with decreasing size and more specifically is inversely proportional to the side length of the square window. It also increases proportionally with the membrane thickness. Pressures > 60 bars could be achieved for 100 nm-thick square Si3N4 windows that are smaller than 65 µm. However, above a certain pressure, the failure of the Si wafer becomes the limiting factor. The failure pressure of the Si wafer is sensitive to the wafer thickness. Moreover, the deformation of the Si3N4 membrane is quantified using vertical scanning interferometry. As an example of the performance of the high-pressure liquid cell optimized for total-fluorescence detected soft x-ray absorption spectroscopy (sXAS), the sXAS spectra at the Ca L edge (∼350 eV) of a CaCl2 aqueous solution are collected under different pressures up to 41 bars.

Full Characterization of the Mechanical Properties of 11-50 nm Ultrathin Films: Influence of Network Connectivity on the Poisson's Ratio.

Precise characterization of the mechanical properties of ultrathin films is of paramount importance for both a fundamental understanding of nanoscale materials and for continued scaling and improvement of nanotechnology. In this work, we use coherent extreme ultraviolet beams to characterize the full elastic tensor of isotropic ultrathin films down to 11 nm in thickness. We simultaneously extract the Young's modulus and Poisson's ratio of low-k a-SiC:H films with varying degrees of hardness and average network connectivity in a single measurement. Contrary to past assumptions, we find that the Poisson's ratio of such films is not constant but rather can significantly increase from 0.25 to >0.4 for a network connectivity below a critical value of ∼2.5. Physically, the strong hydrogenation required to decrease the dielectric constant k results in bond breaking, lowering the network connectivity, and Young's modulus of the material but also decreases the compressibility of the film. This new understanding of ultrathin films demonstrates that coherent EUV beams present a new nanometrology capability that can probe a wide range of novel complex materials not accessible using traditional approaches.

Three-dimensional nanoscale molecular imaging by extreme ultraviolet laser ablation mass spectrometry.

Analytical probes capable of mapping molecular composition at the nanoscale are of critical importance to materials research, biology and medicine. Mass spectral imaging makes it possible to visualize the spatial organization of multiple molecular components at a sample's surface. However, it is challenging for mass spectral imaging to map molecular composition in three dimensions (3D) with submicron resolution. Here we describe a mass spectral imaging method that exploits the high 3D localization of absorbed extreme ultraviolet laser light and its fundamentally distinct interaction with matter to determine molecular composition from a volume as small as 50 zl in a single laser shot. Molecular imaging with a lateral resolution of 75 nm and a depth resolution of 20 nm is demonstrated. These results open opportunities to visualize chemical composition and chemical changes in 3D at the nanoscale.