Search

International benchmarking competitions have become fundamental for the comparative performance assessment of image analysis methods. However, little attention has been given to investigating what can be learnt from these competitions. Do they really generate scientific progress? What are common and successful participation strategies? What makes a solution superior to a competing method? To address this gap in the literature, we performed a multi-center study with all 80 competitions that were conducted in the scope of IEEE ISBI 2021 and MICCAI 2021. Statistical analyses performed based on comprehensive descriptions of the submitted algorithms linked to their rank as well as the underlying participation strategies revealed common characteristics of winning solutions. These typically include the use of multi-task learning (63%) and/or multi-stage pipelines (61%), and a focus on augmentation (100%), image preprocessing (97%), data curation (79%), and postprocessing (66%). The "typical" lead of a winning team is a computer scientist with a doctoral degree, five years of experience in biomedical image analysis, and four years of experience in deep learning. Two core general development strategies stood out for highly-ranked teams: the reflection of the metrics in the method design and the focus on analyzing and handling failure cases. According to the organizers, 43% of the winning algorithms exceeded the state of the art but only 11% completely solved the respective domain problem. The insights of our study could help researchers (1) improve algorithm development strategies when approaching new problems, and (2) focus on open research questions revealed by this work., Comment: accepted to CVPR 2023

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

As climate change poses new and more unpredictable challenges to society, insurance is an essential avenue to protect against loss caused by extreme events. Traditional insurance risk models employ statistical analyses that are inaccurate and are becoming increasingly flawed as climate change renders weather more erratic and extreme. Data-driven parametric insurance could provide necessary protection to supplement traditional insurance. We use a technique referred to as the deep sigma point process, which is one of the Bayesian neural network approaches, for the data analysis portion of parametric insurance using residential internet connectivity dropout in US as a case study. We show that our model has significantly improved accuracy compared to traditional statistical models. We further demonstrate that each state in US has a unique weather factor that primarily influences dropout rates and that by combining multiple weather factors we can build highly accurate risk models for parametric insurance. We expect that our method can be applied to many types of risk to build parametric insurance options, particularly as climate change makes risk modeling more challenging.

Under conditions of strong scattering, a dilemma often arises regarding the best numerical method to use. Main competitors are the Born series, the Beam Propagation Method, and direct solution of the Lippmann-Schwinger equation. However, analytical relationships between the three methods have not yet, to our knowledge, been explicitly stated. Here, we bridge this gap in the literature. In addition to overall insight about aspects of optical scattering that are best numerically captured by each method, our approach allows us to derive approximate error bounds to be expected under various scattering conditions.

We report the importance of anisotropic Coulomb interactions in DFT+U calculations of the electronic and magnetic properties of Mn$_3$O$_4$. The effects of anisotropic interactions in Mn$^{2+}$ and Mn$^{3+}$ are separately examined by defining two different sets of Hubbard parameters: $U^{2+}$ and $J^{2+}$ for Mn$^{2+}$ and $U^{3+}$ and $J^{3+}$ for Mn$^{3+}$. The anisotropic interactions in Mn$^{3+}$ have a significant impact on the physical properties of Mn$_3$O$_4$ including local magnetic moments, canted angle, spontaneous magnetic moment, and superexchange coupling, but those in Mn$^{2+}$ do not make any noticeable difference. Weak ferromagnetic interchain superexchange, observed in experiments, is predicted only if a sizable anisotropic interaction is considered in Mn$^{3+}$. By analyzing the eigenoccupations of the on-site Mn density matrix, we found that the spin channel involving Mn$^{3+}$ $d_{x^2-y^2}$ orbitals, which governs the 90$^\circ$ correlation superexchange, is directly controlled by the anisotropic interactions. These findings demostrate that the exchange correction $J$ for the intraorbital Coulomb potential is of critical importance for first-principles description of reduced Mn oxides containing Mn$^{3+}$ or Mn$^{4+}$.

Solution of the discretized Lippmann-Schwinger equation in the spatial frequency domain involves the inversion of a linear operator specified by the scattering potential. To regularize this inevitably ill-conditioned problem, we propose a machine learning approach: a recurrent neural network with long short-term memory (LSTM) and with the null space projection of the Lippmann-Schwinger kernel on the recurrence path. The learning method is trained using examples of typical scattering potentials and their corresponding scattered fields. We test the proposed method in two cases: electromagnetic scattering by dielectric objects, and electron scattering by multiple screened Coulomb potentials. In both cases the solutions to test examples, disjoint from the training set, were obtained with fewer iterations and were accurate compared to linear solvers. We also observed surprising generalization ability: in the electromagnetic case, an LSTM trained with random arrangements of dielectric spheres was able to obtain the correct solutions for general topologically similar objects, such as polygons. This suggests that the LSTM successfully incorporates the physics of scattering into the inversion algorithm.

SrRuO$_3$ (SRO) films are known to exhibit insulating behavior as their thickness approaches four unit cells. We employ electron energy$-$loss (EEL) spectroscopy to probe the spatially resolved electronic structures of both insulating and conducting SRO to correlate them with the metal$-$insulator transition (MIT). Importantly, the central layer of the ultrathin insulating film exhibits distinct features from the metallic SRO. Moreover, EEL near edge spectra adjacent to the SrTiO$_3$ (STO) substrate or to the capping layer are remarkably similar to those of STO. The site$-$projected density of states based on density functional theory (DFT) partially reflects the characteristics of the spectra of these layers. These results may provide important information on the possible influence of STO on the electronic states of ultrathin SRO., Comment: 6 pages; 5 figures