From What to Why, the Growing Need for a Focus Shift Toward Explainability of AI in Digital Pathology

While it is impossible to deny the performance gains achieved through the incorporation of deep learning (DL) and other artificial intelligence (AI)-based techniques in pathology, minimal work has been done to answer the crucial question of why these algorithms predict what they predict. Tracing back classification decisions to specific input features allows for the quick identification of model bias as well as providing additional information toward understanding underlying biological mechanisms. In digital pathology, increasing the explainability of AI models would have the largest and most immediate impact for the image classification task. In this review, we detail some considerations that should be made in order to develop models with a focus on explainability.

A Study Of Welding Of Heterogenous Polycarbonate Substances Utilizing Hybrid Filaments Of Fused Deposition Modeling

Friction-based welding is one of the most cost-effective and dependable methods for joining thermoplastics. However, there has been minimal work that has demonstrated the procedure/methods/equipment for welding two distinct types of thermoplastics. There is, nevertheless, a significant possibility of connecting the various thermoplastic materials by matching their melt flow index (MFI). One way for modifying the MFI is to reinforce it with micro/nano sized fillers. Fused deposition modelling (FDM) is a fast prototyping technology that employs thermoplastic-based filament to print components. The current study focuses on connecting aluminium (Al) metal powder reinforced acrylonitrile butadiene styrene (ABS) and polyamide 6 (PA6) thermoplastic substrates (3D printed by FDM) utilising friction welding (FW) / friction stir welding (FSW) / friction stir spot welding (FSSW). It was observed that the PA6 with 50% Al fillers (PA6-50% Al) and ABS matrix with 15% Al fillers (ABS-15% Al) produced MFIs of 11.97g/10min and 11.57g/10min, respectively.

Shy Times in 1850s Wellington: will the real architects please stand up?

Wellington was in a period of transition in the 1850s. The first flurry of settlement was easing somewhat and trading was becoming established. However, the earthquakes of 1848 and 1855 shook not only buildings not designed to withstand them, but also the confidence of the immigrant population. People were quick to realise that timber flexed better than brick or cob, but, in the process, they lost several of the earliest buildings with any pretensions to architectural merit. Together with the shaky nature of the economy, and the fact that Auckland was the capital city, there was little incentive for men whose sole training was in architecture to attempt to practice full time.The paucity of architectural records from the 1850s further complicates accurate evaluation of the situation, but it is clear that many of the people designing buildings had multiple skills in several other fields besides architecture. Buildings definitely dated to the 1850s that remain in Wellington can be numbered on one hand and not one of them can be said to have been designed by an architect. The two men with the largest tallies of Wellington building designs in the 1850s also claimed skills in surveying and civil engineering, whereas the two (possibly three) trained architects that we know of seem to have obtained minimal work in their field and to have largely diversified into other occupations. A further five names are associated with Wellington architecture in some way during the 1850s, either with the design of single buildings or simply advertising their services in local newspapers - with no evidence they actually obtained any work. In this paper we look at the backgrounds of the major designers including the trained architects, their work and a few of the factors which caused most of them to seek alternative employment.

Day-Ahead Forecasting of Losses in the Distribution Network

Utility companies in the Nordics have to nominate how much electricity is expected to be lost in their power grid the next day. We present a commercially deployed machine learning system that automates this day-ahead nomination of the expected grid loss. It meets several practical constraints and issues related to, among other things, delayed, missing and incorrect data and a small data set. The system incorporates a total of 24 different models that performs forecasts for three sub-grids. Each day one model is selected for making the hourly day-ahead forecasts for each sub-grid. The deployed system reduced the mean average percentage error (MAPE) with 40% from 12.17 to 7.26 per hour from mid-July to mid-October, 2019. It is robust, flexible and reduces manual work. Recently, the system was deployed to forecast and nominate grid losses for two new grids belonging to a new customer. As the presented system is modular and adaptive, the integration was quick and needed minimal work. We have shared the grid loss data-set on Kaggle.

The Wheel of Change Moves On: Assessing the Severity of Stalking Behavior

Stalking is a significant social issue. The inconsistency as to what defines stalking has resulted in the creation of different methods to measure the crime. However, there has been minimal work done that assesses the severity of individual stalking behaviors. The aim of the present study was to assess the level of stalking behavior in terms of severity within a randomly selected sample of 924 cases from the database of the National Stalking Helpline. Item response theory analyses were used to assist in developing a scale that displays the ranking order of each stalking behavior. These analyses were also used to examine whether the stalking behavioral items created a single continuum of severity of stalking. Results indicated that 16 stalking behavioral items of the 28 items present in the National Stalking Helpline, best represented the severity of stalking. Unwanted communication behaviors such as text messages and phone calls were located at the lower end of the severity scale, whereas criminal damage and death threats were mapped on the higher end of the continuum. The findings also revealed that the 16 items categorized under 6 factors. The findings of the present study provide many implications for stalking agency professionals and criminal justice responses.

A Hierarchical Multicast Key Distribution Protocol

In secure group communication, group keys (GK) are used to ensure the confidentiality of communication. The group key distribution (GKD) is responsible for updating and distributing new group keys when the group membership changes. Most well-known GKD protocols are based on a logical key hierarchy (LKH), where only one group controller (GC) is used. These protocols have various issues, including a single point of failure, meaning that the GC often has a huge workload and can be easily overwhelmed. In this paper, we propose a hierarchical multicast key distribution protocol that supports multi-level controllers to manage a group. Each controller just needs to manage the next-level nodes, and if one fails, the superior controller can replace it with minimal work. The proposed protocol effectively balances the work of controllers, greatly improves the reliability of the group key distribution, and also allows group members to build dynamic conferences without controllers. We provide a security proof of the proposed protocol in a symbolic security model and compare it to other protocols in terms of efficiency, functionality, and security.

Green Synthesis of Symmetric Dimaleamic Acids from Dianilines and Maleic Anhydride: Behind New Bidentate Ligands for MOFs

We herein report the synthesis and characterization of six α,β-unsaturated dicarboxylic acid ligands with different phenyl spacers, and two ligands with a biphenyl and anthraquinone spacers. All these dimaleamic acids were synthesized in 16 to 99% yields via a base-catalyzed maleimide ring opening in water (ligand 2), or by a di-N-acylation from the corresponding diamines and maleic anhydride in acetic acid (ligands 4, 6, 8, 10, 12, 14 and 16). These reactions were performed using green solvents, while requiring minimal work up procedures, making them suitable alternatives to access these types of bidentate ligands quickly, which can be used to fabricate new metal-organic frameworks (MOFs).

A comparison of estimates of global carbon dioxide emissions from fossil carbon sources

Since the first estimate of global CO2 emissions was published in 1894, important progress has been made in the development of estimation methods while the number of available datasets has grown. The existence of parallel efforts should lead to improved accuracy and understanding of emissions estimates, but there remains significant deviation between estimates and relatively poor understanding of the reasons for this. Here I describe the most important global emissions datasets available today and – by way of global, large-emitter, and case examples – quantitatively compare their estimates, exploring the reasons for differences. In many cases differences in emissions come down to differences in system boundaries: which emissions sources are included and which are omitted. With minimal work in harmonising these system boundaries across datasets, the range of estimates of global emissions drops to 5 %, and further work on harmonisation would likely result in an even lower range, without changing the data. Some potential errors were found, and some discrepancies remain unexplained, but it is shown to be inappropriate to conclude that uncertainty in emissions is high simply because estimates exhibit a wide range. While “true” emissions cannot be known, by comparing different datasets methodically, differences that result from system boundaries and allocation approaches can be highlighted and set aside to enable identification of true differences, and potential errors. This must be an important way forward in improving global datasets of CO2 emissions. Data used to generate Figs. 3–18 are available at https://doi.org/10.5281/zenodo.3687042 (Andrew, 2020).

A comparison of estimates of global carbon dioxide emissions from fossil carbon sources

Since the first estimate of global CO2 emissions was published in 1894, important progress has been made in the development of estimation methods while the number of available datasets has grown. The existence of parallel efforts should lead to improved accuracy and understanding of emissions estimates, but there remains significant deviation between estimates and relatively poor understanding of the reasons for this. Here I describe the most important global emissions datasets available today and – by way of global, large-emitter, and case examples – quantitatively compare their estimates, exploring the reasons for differences. In many cases differences in emissions come down to differences in system boundaries: which emissions sources are included and which are omitted. With minimal work in harmonising these system boundaries across datasets, the range of estimates of global emissions drops to 5 %, and further work on harmonisation would likely result in an even lower range, without changing the data. Some potential errors were found, and some discrepancies remain unexplained, but it is shown to be inappropriate to conclude that uncertainty in emissions is high simply because estimates exhibit a wide range. While true emissions cannot be known, by comparing different datasets methodically, differences that result from system boundaries and allocation approaches can be highlighted and set aside to enable identification of true differences, and potential errors. This must be an important way forward in improving global datasets of CO2 emissions. Data used to generate figures 4–19 are available at https://doi.org/10.5281/zenodo.3687042 (Andrew, 2020).