Search

Transforming raw EHR data into machine learning model-ready inputs requires considerable effort. One widely used EHR database is Medical Information Mart for Intensive Care (MIMIC). Prior work on MIMIC-III cannot query the updated and improved MIMIC-IV version. Besides, the need to use multicenter datasets further highlights the challenge of EHR data extraction. Therefore, we developed an extraction pipeline that works on both MIMIC-IV and eICU Collaborative Research Database and allows for model cross validation using these 2 databases. Under the default choices, the pipeline extracted 38766 and 126448 ICU records for MIMIC-IV and eICU, respectively. Using the extracted time-dependent variables, we compared the Area Under the Curve (AUC) performance with prior works on clinically relevant tasks such as in-hospital mortality prediction. METRE achieved comparable performance with AUC 0.723- 0.888 across all tasks. Additionally, when we evaluated the model directly on MIMIC-IV data using a model trained on eICU, we observed that the AUC change can be as small as +0.019 or -0.015. Our open-source pipeline transforms MIMIC-IV and eICU into structured data frames and allows researchers to perform model training and testing using data collected from different institutions, which is of critical importance for model deployment under clinical contexts.

Deep learning models have demonstrated great potential in medical 3D imaging, but their development is limited by the expensive, large volume of annotated data required. Active learning (AL) addresses this by training a model on a subset of the most informative data samples without compromising performance. We compared different AL strategies and propose a framework that minimizes the amount of data needed for state-of-the-art performance. 638 multi-institutional brain tumor MRI images were used to train a 3D U-net model and compare AL strategies. We investigated uncertainty sampling, annotation redundancy restriction, and initial dataset selection techniques. Uncertainty estimation techniques including Bayesian estimation with dropout, bootstrapping, and margins sampling were compared to random query. Strategies to avoid annotation redundancy by removing similar images within the to-be-annotated subset were considered as well. We determined the minimum amount of data necessary to achieve similar performance to the model trained on the full dataset ({\alpha} = 0.1). A variance-based selection strategy using radiomics to identify the initial training dataset is also proposed. Bayesian approximation with dropout at training and testing showed similar results to that of the full data model with less than 20% of the training data (p=0.293) compared to random query achieving similar performance at 56.5% of the training data (p=0.814). Annotation redundancy restriction techniques achieved state-of-the-art performance at approximately 40%-50% of the training data. Radiomics dataset initialization had higher Dice with initial dataset sizes of 20 and 80 images, but improvements were not significant. In conclusion, we investigated various AL strategies with dropout uncertainty estimation achieving state-of-the-art performance with the least annotated data., Comment: 22 pages, 3 figures, 3 tables, 1 supplementary data document. Submitted to Medical Physics in Jan 2023

Although synthetic aperture imaging (SAI) can achieve the seeing-through effect by blurring out off-focus foreground occlusions while recovering in-focus occluded scenes from multi-view images, its performance is often deteriorated by dense occlusions and extreme lighting conditions. To address the problem, this paper presents an Event-based SAI (E-SAI) method by relying on the asynchronous events with extremely low latency and high dynamic range acquired by an event camera. Specifically, the collected events are first refocused by a Refocus-Net module to align in-focus events while scattering out off-focus ones. Following that, a hybrid network composed of spiking neural networks (SNNs) and convolutional neural networks (CNNs) is proposed to encode the spatio-temporal information from the refocused events and reconstruct a visual image of the occluded targets. Extensive experiments demonstrate that our proposed E-SAI method can achieve remarkable performance in dealing with very dense occlusions and extreme lighting conditions and produce high-quality images from pure events. Codes and datasets are available at https://dvs-whu.cn/projects/esai/., Comment: Accepted by IEEE TPAMI. arXiv admin note: text overlap with arXiv:2103.02376

Despite the huge advancement in knowledge discovery and data mining techniques, the X-ray diffraction (XRD) analysis process has mostly remained untouched and still involves manual investigation, comparison, and verification. Due to the large volume of XRD samples from high-throughput XRD experiments, it has become impossible for domain scientists to process them manually. Recently, they have started leveraging standard clustering techniques, to reduce the XRD pattern representations requiring manual efforts for labeling and verification. Nevertheless, these standard clustering techniques do not handle problem-specific aspects such as peak shifting, adjacent peaks, background noise, and mixed phases; hence, resulting in incorrect composition-phase diagrams that complicate further steps. Here, we leverage data mining techniques along with domain expertise to handle these issues. In this paper, we introduce an incremental phase mapping approach based on binary peak representations using a new threshold based fuzzy dissimilarity measure. The proposed approach first applies an incremental phase computation algorithm on discrete binary peak representation of XRD samples, followed by hierarchical clustering or manual merging of similar pure phases to obtain the final composition-phase diagram. We evaluate our method on the composition space of two ternary alloy systems- Co-Ni-Ta and Co-Ti-Ta. Our results are verified by domain scientists and closely resembles the manually computed ground-truth composition-phase diagrams. The proposed approach takes us closer towards achieving the goal of complete end-to-end automated XRD analysis., Comment: Accepted and presented at the International Workshop on Domain-Driven Data Mining (DDDM) as a part of the SIAM International Conference on Data Mining (SDM 2021). Contains 11 pages and 5 figures

We propose an end-to-end music mixing style transfer system that converts the mixing style of an input multitrack to that of a reference song. This is achieved with an encoder pre-trained with a contrastive objective to extract only audio effects related information from a reference music recording. All our models are trained in a self-supervised manner from an already-processed wet multitrack dataset with an effective data preprocessing method that alleviates the data scarcity of obtaining unprocessed dry data. We analyze the proposed encoder for the disentanglement capability of audio effects and also validate its performance for mixing style transfer through both objective and subjective evaluations. From the results, we show the proposed system not only converts the mixing style of multitrack audio close to a reference but is also robust with mixture-wise style transfer upon using a music source separation model.

The volumetric stretch energy has been widely applied to the computation of volume-/mass-preserving parameterizations of simply connected tetrahedral mesh models. However, this approach still lacks theoretical support. In this paper, we provide the theoretical foundation for volumetric stretch energy minimization (VSEM) to compute volume-/mass-preserving parameterizations. In addition, we develop an associated efficient VSEM algorithm with guaranteed asymptotic R-linear convergence. Furthermore, based on the VSEM algorithm, we propose a projected gradient method for the computation of the volume/mass-preserving optimal mass transport map with a guaranteed convergence rate of $\mathcal{O}(1/m)$, and combined with Nesterov-based acceleration, the guaranteed convergence rate becomes $\mathcal{O}(1/m^2)$. Numerical experiments are presented to justify the theoretical convergence behavior for various examples drawn from known benchmark models. Moreover, these numerical experiments show the effectiveness and accuracy of the proposed algorithm, particularly in the processing of 3D medical MRI brain images., Comment: 26 pages

Spin qubits based on silicon metal-oxide semiconductor (Si-MOS) quantum dots (QDs) are promising platforms for large-scale quantum computers. To control spin qubits in QDs, electric dipole spin resonance (EDSR) has been most commonly used in recent years. By delocalizing an electron across a double quantum dots charge state, flopping-mode EDSR has been realized in Si/SiGe QDs. Here, we demonstrate a flopping-mode spin qubit in a Si-MOS QD via Elzerman single-shot readout. When changing the detuning with a fixed drive power, we achieve s-shape spin resonance frequencies, an order of magnitude improvement in the spin Rabi frequencies, and virtually constant spin dephasing times. Our results offer a route to large-scale spin qubit systems with higher control fidelity in Si-MOS QDs., Comment: 5 pages, 4 figures

Music mixing traditionally involves recording instruments in the form of clean, individual tracks and blending them into a final mixture using audio effects and expert knowledge (e.g., a mixing engineer). The automation of music production tasks has become an emerging field in recent years, where rule-based methods and machine learning approaches have been explored. Nevertheless, the lack of dry or clean instrument recordings limits the performance of such models, which is still far from professional human-made mixes. We explore whether we can use out-of-domain data such as wet or processed multitrack music recordings and repurpose it to train supervised deep learning models that can bridge the current gap in automatic mixing quality. To achieve this we propose a novel data preprocessing method that allows the models to perform automatic music mixing. We also redesigned a listening test method for evaluating music mixing systems. We validate our results through such subjective tests using highly experienced mixing engineers as participants., Comment: 23rd International Society for Music Information Retrieval Conference (ISMIR), December, 2022. Source code, demo and audio examples: https://marco-martinez-sony.github.io/FxNorm-automix/ - added acknowledgements

Interfacial Dzyaloshinskii-Moriya interaction (DMI) has long been observed in normal metal/ferromagnetic multilayers, enabling the formation of chiral domain walls, skyrmions and other 2D antisymmetric spin textures confined within a single ferromagnetic layer, while more recent works on interlayer DMI reveal new pathways in realizing novel chiral 3D spin textures between two separate layers.Here, we report on interlayer DMI between two orthogonally magnetized ferromagnetic layers (CoFeB/Co) mediated by a Pt layer, and confirm the chiral nature of the observed effective field of up to 37 Oe through asymmetric hysteresis loops under in-plane field.We highlight the importance of growth-induced in-plane symmetry breaking, resulting in a sizable interlayer DMI and a universal characteristic vector through wedge deposition of the samples.We further perform deterministic current-driven magnetization switching in the perpendicularly magnetized Co layer utilizing solely the effective field from the interlayer DMI.These results demonstrate the potential of interlayer DMI in facilitating deterministic field-free switching in spin memory applications., Comment: 22 pages, 3 figures

In this paper, we first extend the finite distortion problem from the bounded domains in $\mathbb{R}^2$ to the closed genus-zero surfaces in $\mathbb{R}^3$ by the stereographic projection. Then we derive a theoretical foundation for spherical equiareal parameterizations between a simply connected closed surface $\mathcal{M}$ and a unit sphere $\mathbb{S}^2$ via minimizing the total area distortion energy on $\overline{\mathbb{C}}$. Provided we determine the minimizer of the total area distortion energy, the minimizer composed with the initial conformal map determines the equiareal map between the extended planes. Taking the inverse stereographic projection, we can derive the equiareal map between $\mathcal{M}$ and $\mathbb{S}^2$. The total area distortion energy can be rewritten into the sum of Dirichlet energies associated with the southern and northern hemispheres, respectively, and can be decreased by alternatingly solving the corresponding Laplacian equations. Based on this foundational theory, we develop a modified stretch energy minimization for the computation of the spherical equiareal parameterization between $\mathcal{M}$ and $\mathbb{S}^2$. In addition, under some mild conditions, we verify that our proposed algorithm has asymptotically R-linear convergence or forms a quasi-periodic solution. Numerical experiments on various benchmarks validate the assumptions for convergence always hold and indicate the efficiency, reliability and robustness of the developed modified stretch energy minimization.

In this paper, we first derive a theoretical basis for spherical conformal parameterizations between a simply connected closed surface $\mathcal{S}$ and a unit sphere $\mathbb{S}^2$ by minimizing the Dirichlet energy on $\overline{\mathbb{C}}$ by stereographic projection. The Dirichlet energy can be rewritten as the sum of the energies associated with the southern and northern hemispheres and can be decreased under an equivalence relation by alternatingly solving the corresponding Laplacian equations. Based on this theoretical foundation, we develop a modified Dirichlet energy minimization with nonequivalence deflation for the computation of the spherical conformal parameterization between $\mathcal{S}$ and $\mathbb{S}^2$. In addition, under some mild conditions, we verify the asymptotically R-linear convergence of the proposed algorithm. Numerical experiments on various benchmarks confirm that the assumptions for convergence always hold and indicate the efficiency, reliability and robustness of the developed modified Dirichlet energy minimization., Comment: 29 pages

Spin-orbit torque characterizations on magnetic heterostructures with perpendicular anisotropy are demonstrated on a projected vector field magnet via hysteresis loop shift measurement and harmonic Hall measurement with planar Hall correction. Accurate magnetic field calibration of the vector magnet is realized with the help of deep learning models, which are able to capture the nonlinear behavior between the generated magnetic field and the currents applied to the magnet. The trained models can successfully predict the applied current combinations under the circumstances of magnetic field scans, angle scans, and hysteresis loop shift measurements. The validity of the models is further verified, complemented by the comparison of the spin-orbit torque characterization results obtained from the deep-learning-trained vector magnet system with those obtained from a conventional setup comprised of two separated electromagnets. The damping-like spin-orbit torque (DL-SOT) efficiencies (|$\xi_{DL}$|) extracted from the vector magnet and the traditional measurement configuration are consistent, where |$\xi_{DL}$| $\approx$ 0.22 for amorphous W and |$\xi_{DL}$| $\approx$ 0.02 for $\alpha$-W. Our work provides an advanced method to meticulously control a vector magnet and to conveniently perform various spin-orbit torque characterizations.

In High Energy Physics (HEP), experimentalists generate large volumes of data that, when analyzed, helps us better understand the fundamental particles and their interactions. This data is often captured in many files of small size, creating a data management challenge for scientists. In order to better facilitate data management, transfer, and analysis on large scale platforms, it is advantageous to aggregate data further into a smaller number of larger files. However, this translation process can consume significant time and resources, and if performed incorrectly the resulting aggregated files can be inefficient for highly parallel access during analysis on large scale platforms. In this paper, we present our case study on parallel I/O strategies and HDF5 features for reducing data aggregation time, making effective use of compression, and ensuring efficient access to the resulting data during analysis at scale. We focus on NOvA detector data in this case study, a large-scale HEP experiment generating many terabytes of data. The lessons learned from our case study inform the handling of similar datasets, thus expanding community knowledge related to this common data management task.

Current-induced spin-orbit torque (SOT) is regarded as a promising mechanism for driving neuromorphic behavior in spin-orbitronic devices. In principle, the strong SOT in heavy metal-based magnetic heterostructure is attributed to the spin-orbit coupling (SOC)-induced spin Hall effect (SHE) and/or the spin Rashba-Edelstein effect (SREE). Recently, SOC-free mechanisms such as the orbital angular momentum (OAM)-induced orbital Hall effect (OHE) and/or the orbital Rashba-Edelstein effect (OREE) have been proposed to generate sizable torques comparable to those from the conventional spin Hall mechanism. In this work, we show that the orbital current can be effectively generated by the nitrided light metal Cu. The overall damping-like SOT efficiency, which consists of both the spin and the orbital current contributions, can be tailored from ~ 0.06 to 0.4 in a Pt/Co/CuNx magnetic heterostructure by tuning the nitrogen doping concentration. Current-induced magnetization switching further verifies the efficacy of such orbital current with a critical switching current density as low as Jc ~ 5 x 10^10 A/m2. Most importantly, the orbital-current-mediated memristive switching behavior can be observed in such heterostructures, which reveals that the gigantic SOT and efficient magnetization switching are the tradeoffs for the applicable window of memristive switching. Our work provides insights into the role of orbital current might play in SOT neuromorphic devices and paves a new route for making energy-efficient spin-orbitronic devices.

We propose Path-CNN, a method for the segmentation of centerlines of tubular structures by embedding convolutional neural networks (CNNs) into the progressive minimal path method. Minimal path methods are widely used for topology-aware centerline segmentation, but usually these methods rely on weak, hand-tuned image features. In contrast, CNNs use strong image features which are learned automatically from images. But CNNs usually do not take the topology of the results into account, and often require a large amount of annotations for training. We integrate CNNs into the minimal path method, so that both techniques benefit from each other: CNNs employ learned image features to improve the determination of minimal paths, while the minimal path method ensures the correct topology of the segmented centerlines, provides strong geometric priors to increase the performance of CNNs, and reduces the amount of annotations for the training of CNNs significantly. Our method has lower hardware requirements than many recent methods. Qualitative and quantitative comparison with other methods shows that Path-CNN achieves better performance, especially when dealing with tubular structures with complex shapes in challenging environments., Comment: Accepted to IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, 2022

Additive manufacturing has been vastly applied to fabricate various structures of nickel-titanium (NiTi) shape memory alloys due to its flexibility to create complex structures with minimal defects. However, the microstructure heterogeneity and secondary phase formation are two main problems that impede the further application of NiTi alloys. Although post-heat treatment is usually adopted to improve or manipulate NiTi alloy properties, it cannot realize the spatial control of thermal and/or mechanical properties of NiTi alloys. To overcome the limitations of uniform post-heat treatment, this study proposes an in-situ heat treatment strategy that is integrated into the directed energy deposition of NiTi alloys. The proposed method will potentially lead to new manufacturing capabilities to achieve location-dependent performance or property manipulation. The influences of in-situ heat treatment on the thermal and mechanical properties of printed NiTi structures were investigated. The investigations were carried out in terms of thermal cycling, microstructure evolution, and mechanical properties by 3D finite element simulations and experimental characterizations. A low-power laser beam was adopted to localize the in-situ heat treatment only to the current printed layer, facilitating a reverse peritectic reaction and a transient high solution treatment successively. The proposed in-situ heat treatment on the specimen results in a more obvious phase transformation peak in the differential scanning calorimetry curves, about 50%~70% volume reduction for the Ti2Ni phase, and approximately 35 HV reduction on microhardness.