Diversity Branches Research Articles

S2DA-Net: Spatial and spectral-learning double-branch aggregation network for liver tumor segmentation in CT images

Accurate liver tumor segmentation is crucial for aiding radiologists in hepatocellular carcinoma evaluation and surgical planning. While convolutional neural networks (CNNs) have been successful in medical image segmentation, they face challenges in capturing long-term dependencies among pixels. On the other hand, Transformer-based models demand a high number of parameters and involve significant computational costs. To address these issues, we propose the Spatial and Spectral-learning Double-branched Aggregation Network (S2DA-Net) for liver tumor segmentation. S2DA-Net consists of a double-branched encoder and a decoder with a Group Multi-Head Cross-Attention Aggregation (GMCA) module, Two branches in the encoder consist of a Fourier Spectral-learning Multi-scale Fusion (FSMF) branch and a Multi-axis Aggregation Hadamard Attention (MAHA) branch. The FSMF branch employs a Fourier-based network to learn amplitude and phase information, capturing richer features and detailed information without introducing an excessive number of parameters. The FSMF branch utilizes a Fourier-based network to capture amplitude and phase information, enriching features without introducing excessive parameters. The MAHA branch incorporates spatial information, enhancing discriminative features while minimizing computational costs. In the decoding path, a GMCA module extracts local information and establishes long-term dependencies, improving localization capabilities by amalgamating features from diverse branches. Experimental results on the public LiTS2017 liver tumor datasets show that the proposed segmentation model achieves significant improvements compared to the state-of-the-art methods, obtaining dice per case (DPC) 69.4 % and global dice (DG) 80.0 % for liver tumor segmentation on the LiTS2017 dataset. Meanwhile, the pre-trained model based on the LiTS2017 datasets obtain, DPC 73.4 % and an DG 82.2 % on the 3DIRCADb dataset.

HICL: Hierarchical Intent Contrastive Learning for sequential recommendation

Users’ diverse intentions drive the subsequent interaction based on users’ dynamic behavior trajectories. Despite the impressive progress of existing work on capturing the user’s intent, these methods still have the following challenges: (1) How can the intent effectively be exploited in sparse and long-distance interacted items? (2) How to improve the performance of a recommendation system by fusing spatial and temporal information? To further exploit the intent from local and global contexts, we present a novel recommendation framework, namely hierarchical intent Contrastive Learning (HICL) for SR. First, a graph encoder is employed to enhance item embedding by incorporating global context information. And then, the sequential encoder is leveraged as a prediction baseline from a historical perspective. Moreover, diverse intent Contrastive Learning branches are integrated to model user latent by utilizing a new loss function. Specifically, sequence contrastive learning (CL) employs sequence augment to model the users’ local intent, while graph CL employs graph augment and feature clustering to model the users’ global latent from semantic and structural graph views. Extensive experiments are conducted on four real-world datasets, including highly sparse datasets, and the experiment results demonstrate the superiority of our model over the state-of-the-art.

A novel interactive deep cascade spectral graph convolutional network with multi-relational graphs for disease prediction

Graph neural networks (GNNs) have recently grown in popularity for disease prediction. Existing GNN-based methods primarily build the graph topological structure around a single modality and combine it with other modalities to acquire feature representations of acquisitions. The complicated relationship in each modality, however, may not be well highlighted due to its specificity. Further, relatively shallow networks restrict adequate extraction of high-level features, affecting disease prediction performance. Accordingly, this paper develops a new interactive deep cascade spectral graph convolutional network with multi-relational graphs (IDCGN) for disease prediction tasks. Its crucial points lie in constructing multiple relational graphs and dual cascade spectral graph convolution branches with interaction (DCSGBI). Specifically, the former designs a pairwise imaging-based edge generator and a pairwise non-imaging-based edge generator from different modalities by devising two learnable networks, which adaptively capture graph structures and provide various views of the same acquisition to aid in disease diagnosis. Again, DCSGBI is established to enrich high-level semantic information and low-level details of disease data. It devises a cascade spectral graph convolution operator for each branch and incorporates the interaction strategy between different branches into the network, successfully forming a deep model and capturing complementary information from diverse branches. In this manner, more favorable and sufficient features are learned for a reliable diagnosis. Experiments on several disease datasets reveal that IDCGN exceeds state-of-the-art models and achieves promising results.

MedBench: A Large-Scale Chinese Benchmark for Evaluating Medical Large Language Models

The emergence of various medical large language models (LLMs) in the medical domain has highlighted the need for unified evaluation standards, as manual evaluation of LLMs proves to be time-consuming and labor-intensive. To address this issue, we introduce MedBench, a comprehensive benchmark for the Chinese medical domain, comprising 40,041 questions sourced from authentic examination exercises and medical reports of diverse branches of medicine. In particular, this benchmark is composed of four key components: the Chinese Medical Licensing Examination, the Resident Standardization Training Examination, the Doctor In-Charge Qualification Examination, and real-world clinic cases encompassing examinations, diagnoses, and treatments. MedBench replicates the educational progression and clinical practice experiences of doctors in Mainland China, thereby establish- ing itself as a credible benchmark for assessing the mastery of knowledge and reasoning abilities in medical language learning models. We perform extensive experiments and conduct an in-depth analysis from diverse perspectives, which culminate in the following findings: (1) Chinese medical LLMs underperform on this benchmark, highlighting the need for significant advances in clinical knowledge and diagnostic precision. (2) Several general-domain LLMs surprisingly possess considerable medical knowledge. These findings elucidate both the capabilities and limitations of LLMs within the context of MedBench, with the ultimate goal of aiding the medical research community.

Extracellular vesicle formation in Euryarchaeota is driven by a small GTPase

Since their discovery, extracellular vesicles (EVs) have changed our view on how organisms interact with their extracellular world. EVs are able to traffic a diverse array of molecules across different species and even domains, facilitating numerous functions. In this study, we investigate EV production in Euryarchaeota, using the model organism Haloferax volcanii. We uncover that EVs enclose RNA, with specific transcripts preferentially enriched, including those with regulatory potential, and conclude that EVs can act as an RNA communication system between haloarchaea. We demonstrate the key role of an EV-associated small GTPase for EV formation in H. volcanii that is also present across other diverse evolutionary branches of Archaea. We propose the name, ArvA, for the identified family of archaeal vesiculating GTPases. Additionally, we show that two genes in the same operon with arvA (arvB and arvC) are also involved in EV formation. Both, arvB and arvC, are closely associated with arvA in the majority of other archaea encoding ArvA. Our work demonstrates that small GTPases involved in membrane deformation and vesiculation, ubiquitous in Eukaryotes, are also present in Archaea and are widely distributed across diverse archaeal phyla.

Multiple Branch Deep Neural Network and Feature Fusion Model for Underwater Image Enhancement

Underwater imaging faces formidable challenges due to light absorption, scattering, and restricted visibility, demanding sophisticated enhancement methods. This research presents a novel Multiple Branch Deep Neural Network and Feature Fusion (MB-DNN-FF) Model tailored specifically for Underwater Images Enhancement (UWIE), with a focus on its application to the UFO120 dataset. Driven by the limitations of existing techniques, our model harnesses the potential of Deep Learning (DL) and Feature Fusion (FF) to effectively address the intricate complexities present in underwater environments. The innovative architecture incorporates multiple branches, each strategically designed to tackle distinct challenges such as contrast degradation, limited visibility, and color distortion. Central to our model is feature fusion, a critical aspect that harmoniously integrates information from diverse branches, thereby enhancing overall image quality. The training and optimization processes are thoroughly detailed, encompassing unique strategies and loss functions fine-tuned for the nuances of UWIE, with emphasis on the characteristics of the UFO120 dataset. Experimental evaluations leverage the comprehensive UFO120 dataset, employing established performance metrics to quantitatively assess the efficacy of our model. Results exhibit substantial improvements over baseline models and state-of-the-art methods, showcasing the effectiveness and versatility of the proposed multiple branch model, particularly on the UFO120 dataset. The discussion interprets these findings in the context of underwater imaging challenges, highlighting the model's effectiveness and outlining potential applications. This research contributes a valuable asset to the area of UWIE, offering a nuanced and potent solution through the integration of multiple branches and feature fusion, validated on the UFO120 dataset. The conclusions underscore the significance of the MB-DNN-FF in advancing the state-of-the-art in UWIE, opening avenues for future research and practical applications in marine sciences, surveillance, and exploration.

Diverse branch feature refinement network for efficient multi‐scale super‐resolution

AbstractDespite the existence of various super‐resolution (SR) methods, most of them focus on designing models for specific upscaling factors rather than fully exploiting inter‐scale correlation to improve efficiency. In contrast, multi‐scale SR methods can effectively reduce the redundancy of network parameters by aggregating the feature extraction processes corresponding to multiple scales into a unified process. The aim of this study is to enhance the compactness and efficiency of the SR model. Thus, an efficient multi‐scale SR method called the diverse branch feature refinement network (DBFRN) is proposed. By decoupling the training process and inference process based on the idea of structural re‐parameterization, multi‐branch topology is adopted to enrich multi‐scale learning and merge branches to achieve efficient inference with equivalent effects. Specifically, two re‐parameterization strategies are designed and two corresponding feature refinement blocks for different feature levels in multi‐scale SR network. Extensive experiments demonstrate that the proposed multi‐scale SR method is effective and efficient, and it can outperform advanced single‐scale methods in terms of quantity and quality.

MS3Net: a deep ensemble learning approach for ship classification in heterogeneous remote sensing data

ABSTRACT Maritime ship classification is essential for effectively monitoring oceanic activities but faces challenges when using heterogeneous remote sensing data. This research presents a novel dataset called Heterogeneous Ship Data, created by combining synthetic aperture radar (SAR) and optical satellite images of ships. The integration of multimodal data aims to harness the complementary advantages of SAR’s all-weather imaging capabilities and optical data’s higher interretability and shape discrimination. To leverage the heterogeneous dataset, this paper proposes an ensemble deep learning model called MS3Net (Multi-convolutional network using Spatial attention and Second-order pooling for Ship image classification). MS3Net integrates spatial attention mechanisms and second-order pooling. MS3Net has multiple convolutional branches, each tailored to extract distinct representations from the SAR and optical data through varied filter sizes and kernels. The spatial attention units focus the learning on the most informative parts of the image while suppressing irrelevant features. The second-order pooling captures higher-order interactions between feature elements to model complex relationships. A concatenation fusion strategy aggregates the diverse branch outputs into a joint multimodal feature representation for enhanced discrimination. Extensive experiments demonstrate MS3Net’s superior classification accuracy over prior state-of-the-art approaches. The proposed heterogeneous data representation coupled with the attention-based deep ensemble learning model contributes to maritime surveillance by enabling robust ship recognition across diverse sensing modalities.

Applying optimized YOLOv8 for heritage conservation: enhanced object detection in Jiangnan traditional private gardens

This study aims to promote the protection and inheritance of cultural heritage in private gardens in the Jiangnan area of China. By establishing a precise visual labeling system and accelerating the construction of a database for private garden features, we deepen the understanding of garden design philosophy. To this end, we propose an improved Jiangnan private garden recognition model based on You Only Look Once (YOLO) v8. This model is particularly suitable for processing garden environments with characteristics such as single or complex structures, rich depth of field, and cluttered targets, effectively enhancing the accuracy and efficiency of object recognition. This design integrates the Diverse Branch Block (DBB), Bidirectional Feature Pyramid Network (BiFPN), and Dynamic Head modules (DyHead) to optimize model accuracy, feature fusion, and object detection representational capability, respectively. The enhancements elevated the model's accuracy by 8.7%, achieving a mean average precision (mAP@0.5) value of 57.1%. A specialized dataset, comprising 4890 images and encapsulating various angles and lighting conditions of Jiangnan private gardens, was constructed to realize this. Following manual annotation and the application of diverse data augmentation strategies, the dataset bolsters the generalization and robustness of the model. Experimental outcomes reveal that, compared to its predecessor, the improved model has witnessed increments of 15.16%, 3.25%, and 11.88% in precision, mAP0.5, and mAP0.5:0.95 metrics, respectively, demonstrating exemplary performance in the accuracy and real-time recognition of garden target elements. This research not only furnishes robust technical support for the digitization and intelligent research of Jiangnan private gardens but also provides a potent methodological reference for object detection and classification research in analogous domains.

The art of effective oversight: unravelling the success of Israel's Cabel Inquiry Committee

ABSTRACT Legislative oversight of the executive branch is a fundamental mechanism for ensuring government accountability, good governance, and enhancing democracy. However, formal oversight powers alone are insufficient, and the practical application eventually determines effectiveness. In contemporary times, a distinct set of political phenomena exists, notably the emergence of populism and polarisation, a deep-seated crisis of representation, and a systematic decline in public trust regarding diverse branches of government, coupled with the personalisation of politics and the weakening of political parties. In light of these developments, it is essential to engage in a meticulous examination of the means by which legislatures can enhance their capacity to effectively oversee governments. This article examines the parliamentary committee of inquiry led by Knesset Member Eitan Cabel in Israel from 2017 to 2019, which focused on credit arrangements for large business borrowers. This committee's rigorous investigation into financial system failures makes it an exemplary case of robust oversight in action. The article analyses the Committee establishment, composition, mandate, systematic meeting-planning and questioning process, unique physical setting, report publishing, and other details. The analysis of the Committee's work reveals important insights regarding effective oversight, and isolates the practices which made its work such an augmented performance of legislative oversight. Factors such as the committee's activeness, cooperation, independence, preparation, respectful yet critical approach, transparency, accurate use of media, and precise procedure, created an exceptional performance of legislative oversight. The article concludes that while legislative oversight relies partly on formal mechanisms, it is primarily driven by how political actors apply them, and reveals the importance and centrality of the human factor. This single case analysis, focusing on the committee's oversight practices can inform comparative learning on delivering robust oversight beyond Israel's context, and derive widely applicable insights into key practices for effective legislative supervision of the executive branch.

A Study on Some Topics in Algebra

This study provides a comprehensive examination of various topics in algebra, exploring its historical development, fundamental concepts, and diverse branches. The paper delves into elementary algebra, abstract algebra, linear algebra, and commutative algebra, offering detailed insights into each area. Additionally, it introduces a problem-analysis model to evaluate algebraic concepts. This exploration aims to highlight the significance of algebra in mathematical theory and its practical applications, underscoring its role as a foundational pillar in both academic research and applied sciences.

Infrared Target Detection Based on Interval Sampling Weighting and 3D Attention Head in Complex Scenario

Thermal infrared detection technology can enable night vision and is robust in complex environments, making it highly advantageous for various fields. However, infrared images have low resolution and high noise, resulting in limited detailed information being available about the target object. This difficulty is further amplified when detecting small targets, which are prone to occlusion. In response to these challenges, we propose a model for infrared target detection designed to achieve efficient feature representation. Firstly, an interval sampling weighted (ISW) module is proposed, which strengthens the fusion network’s spatial relationship modeling, thereby elevating the model’s generalization capability across diverse target-density regions. Next, a detection head founded on 3D attention (TAHNet) is introduced, which helps the network more comprehensively understand the feature details of the target. This enhances the accuracy of the model in identifying the target object’s location, reduces false positives and false negatives, and optimizes the network’s performance. Furthermore, to our model, we introduce the C2f module to transfer gradient information across multiple branches. The features learned using diverse branches interact and fuse in subsequent stages, further enhancing the model’s representation ability and understanding of the target. Experimental outcomes validate the efficacy of the proposed model, showcasing state-of-the-art detection performance on FLIR and KAIST thermal infrared datasets and showing strong antiocclusion and robustness in complex scenes.

REGIONAL ECONOMY AND MARKETING OF TERRITORIES: THE METHODOLOGY OF INTERDISCIPLINARY RESEARCH

Interdisciplinary research allows to receive versatile and more profound scientific knowledge about a complex natural phenomenon or an ambiguous event of social life by using approaches and methods of diverse branches of knowledge. The main condition of using these approaches and methods of diverse branches of knowledge in one and the same research is the generality of the object and the subject, the core purpose. For example, studying the motives of labor behavior we may combine approaches and methods of economy, sociology, psychology; studying the causes of agression we may use knowledge in biochemistry of primate brain along with the rest. In every single case the researcher builds a unique and original methodology of research. Quite often interdisciplinary research are made in adjacent fields of knowledge (economic theory, labor economy, management theory). Fields of knowledge that can be called interdisciplinary are based on interdisciplinary research. Regional economy is an interdisciplinary branch of science. Its structure includes knowledge of economy, sociology, geography, geology, etc. In this article the idea of amplification the interdisciplinary branch of science “Regional economy” by methods of marketing the territories is proved. This way of regional econo

Cross-Machine Transfer Fault Diagnosis by Ensemble Weighting Subdomain Adaptation Network

Many domain adaptation models have been explored for fault transfer diagnosis. However, most of them only consider global domain adaptation of two domains while neglecting the fine-grained class-wise distribution alignment between the source and target domains. Thus, these models cannot satisfy the diagnostic requirement in some cases. A new ensemble weighting subdomain adaptation network (EWSAN) diagnostic model is established in this paper to improve the degree of domain confusion. In EWSAN, an enhanced joint distribution alignment (EJDA) mechanism is proposed. A multiscale top classifier with multiple diverse branches is designed based on ensemble learning to better achieve EJDA. Ensemble voting with the multiscale top classifier can obtain more reliable pseudo labels in the EJDA mechanism. An ensemble weighting maximum mean discrepancy (EWMMD) with the class weight is constructed to enhance the fine-grained domain confusion. Moreover, the closed and partial transfer diagnostic tasks are made available. Furthermore, the information entropy is introduced to increase the confidence coefficient of the pseudo label. The proposed EWSAN diagnostic model is evaluated via multiple closed, and partial fault transfer diagnosis experiments cross machines. The experimental results validate its effectiveness and superiority.

Assessing co-diversification in host-associated microbiomes.

When lineages of hosts and microbial symbionts engage in intimate interactions over evolutionary timescales, they can diversify in parallel (i.e., co-diversify), producing associations between the lineages' phylogenetic histories. Tests for co-diversification of individual microbial lineages and their hosts have been developed previously, and these have been applied to discover ancient symbioses in diverse branches of the tree of life. However, most host-microbe relationships are not binary but multipartite, in that a single host-associated microbiota can contain many microbial lineages, generating challenges for assessing co-diversification. Here, we review recent evidence for co-diversification in complex microbiota, highlight the limitations of prior studies, and outline a hypothesis testing approach designed to overcome some of these limitations. We advocate for the use of microbiota-wide scans for co-diversifying symbiont lineages and discuss tools developed for this purpose. Tests for co-diversification for simple host symbiont systems can be extended to entire phylogenies of microbial lineages (e.g., metagenome-assembled or isolate genomes, amplicon sequence variants) sampled from host clades, thereby providing a means for identifying co-diversifying symbionts present within complex microbiota. The relative ages of symbiont clades can corroborate co-diversification, and multi-level permutation tests can account for multiple comparisons and phylogenetic non-independence introduced by repeated sampling of host species. Discovering co-diversifying lineages will generate powerful opportunities for interrogating the molecular evolution and lineage turnover of ancestral, host-species specific symbionts within host-associated microbiota.

Historical sound changes in Central Chadic (Afroasiatic)

Abstract In its entirety as much as with regard to its four branches, the Chadic language family poses challenges to the application of the Neogrammarian-school comparative method, not the least because of the immense time-depth involved and aggravated by certain typological peculiarities of Chadic phonology and morphology as inherited from its Afroasiatic ancestry. This is particularly true for Central Chadic, which—with 80 languages—is the most numerous and most diverse branch of Chadic, which in total counts almost 200 languages and thereby more than half of all known Afroasiatic languages. Both more or less ‘regular’ and ‘sporadic’ sound changes criss-cross the territories currently occupied by speakers of Central Chadic languages in Nigeria, Cameroon, and Chad, which allow confident identification of cognates. Some cognates have survived many millennia of language history practically unchanged with regard to their phonetic realisations, while others differ remarkably from the reconstructed proto-language forms beyond ad hoc recognisability. Recent historical comparative research by Richard Gravina (2014, 2015) and the present author (2022, 2023) involving 66 spoken languages and some 230 cognate lexical items have unearthed much of the linguistic histories behind the massive synchronic diversity of the modern Central Chadic languages. The recent research by the author has added to our understanding of historical Central Chadic phonology by unravelling the phonological processes by which the languages have developed sets of new phonemes, in addition to tracing more or less ‘regular’ and ‘sporadic’ sound changes that the PCC segmental inventories of vowels and consonants underwent. Modern Central Chadic languages show sets of innovative vowels and consonant whose emergence can largely be attributed, besides occasional instances of segmental fusion, to the ‘colouring’ effects of so-called prosodies. This makes the analysis of prosodic features such as palatalisation, labialisation, prenasalisation and glottalisation essential in order to understand the evolution of modern Central Chadic languages.

Space-dependent intermittent feedback can control birhythmicity.

Birhythmicity is evident in many nonlinear systems, which include physical and biological systems. In some living systems, birhythmicity is necessary for response to the varying environment while unnecessary in some physical systems as it limits their efficiency. Therefore, its control is an important area of research. This paper proposes a space-dependent intermittent control scheme capable of controlling birhythmicity in various dynamical systems. We apply the proposed control scheme in five nonlinear systems from diverse branches of natural science and demonstrate that the scheme is efficient enough to control the birhythmic oscillations in all the systems. We derive the analytical condition for controlling birhythmicity by applying harmonic decomposition and energy balance methods in a birhythmic van der Pol oscillator. Further, the efficacy of the control scheme is investigated through numerical and bifurcation analyses in a wide parameter space. Since the proposed control scheme is general and efficient, it may be employed to control birhythmicity in several dynamical systems.

R2R3-MYB transcription factor PhMYB2 positively regulates anthocyanin biosynthesis in Pericallis hybrida

Cineraria (Pericallis hybrida) is an ideal species for anthocyanin accumulation studies in ornamental plants due to its bright flower colors and diverse anthocyanin branching metabolic pathways. However, the regulatory pattern of anthocyanin metabolism was still unclear in previous research on the flower color of cineraria. In order to explore the anthocyanin synthesis mechanism of pure color cultivar of cineraria, we isolated a novel R2R3-MYB gene PhMYB2 based on the previous transcriptome of blue cineraria. PhMYB2 was significantly expressed in carmine and blue cineraria, and its expression profile corresponded with the pattern of anthocyanin accumulation in the ray florets. PhMYB2 could activate the transcription of anthocyanin biosynthetic genes PhF3′H1, PhF3′5′H, PhDFR3, and PhANS when transiently silenced PhMYB2 in cineraria leaves by virus-induced gene silencing (VIGS). Meanwhile, PhMYB2 could interact with PhbHLH17, which belonged to TT8-clade, while PhbHLH17 could interact with a possible R2R3-MYB repressor PhMYB4.Y1H and dual-luciferase assays revealed that PhMYB2 binds and activates the promoter of PhDFR3, and when PhMYB2 interacts with PhbHLH17, the complex can significantly enhance the activity of the PhDFR3 promoter. Furthermore, when PhMYB4 co transformed with PhMYB2 and PhbHLH17, the activity of the PhDFR3 promoter was significantly reduced. Based on this, we speculated that PhMYB4 might fine-tune the anthocyanin synthesis by interacting with PhbHLH17 and competing with PhMYB2. The above results clarified the mechanism of PhMYB2-PhbHLH17 promoting anthocyanin synthesis and improved the anthocyanin regulatory network in cineraria.

Channel Norm-Based Energy Detection With Diversity Reception for Amplitude-Shift Keying in Rayleigh Fading

Optimum and suboptimum energy detection receivers with diversity reception in Rayleigh fading, that only require limited channel state information based on the channel norm, are introduced. The importance of these channel norm based energy detection (CNBED) receivers is that they achieve the same diversity order as that of a coherent receiver. Furthermore, their performance degradation when compared to a coherent receiver can be made sufficiently low by appropriate choice of parameters. Due to their performance and the low complexity of energy detection, the CNBED techniques have potential applications to internet of things. Analysis of the CNBED receivers is performed for multi-level amplitude-shift keying (ASK). For the optimum receiver, an analytical approximation of the symbol error probability (SEP) as a sum of univariate integrals involving averaging of Gaussian Q-functions is derived. For the suboptimum receiver, closed form expressions for the exact SEP and for the approximate SEP under the condition of high signal-to-noise ratio (SNR) are obtained; in addition, using asymptotic approximations, expressions for the optimal ASK transmit levels with high SNR and large number of receive diversity branches, that minimize the SEP subject to an energy constraint, are found. Numerical results are provided to demonstrate the performance of the receivers.

Foundations of Nonequilibrium Statistical Mechanics in Extended State Space

The review provides a pedagogical but comprehensive introduction to the foundations of a recently proposed statistical mechanics (μNEQT) of a stable nonequilibrium thermodynamic body, which may be either isolated or interacting. It is an extension of the well-established equilibrium statistical mechanics by considering microstates mk in an extended state space in which macrostates (obtained by ensemble averaging A^) are uniquely specified so they share many properties of stable equilibrium macrostates. The extension requires an appropriate extended state space, three distinct infinitessimals dα=(d,de,di) operating on various quantities q during a process, and the concept of reduction. The mechanical process quantities (no stochasticity) like macrowork are given by A^dαq, but the stochastic quantities C^αq like macroheat emerge from the commutator C^α of dα and A^. Under the very common assumptions of quasi-additivity and quasi-independence, exchange microquantities deqk such as exchange microwork and microheat become nonfluctuating over mk as will be explained, a fact that does not seem to have been appreciated so far in diverse branches of modern statistical thermodynamics (fluctuation theorems, quantum thermodynamics, stochastic thermodynamics, etc.) that all use exchange quantities. In contrast, dqk and diqk are always fluctuating. There is no analog of the first law for a microstate as the latter is a purely mechanical construct. The second law emerges as a consequence of the stability of the system, and cannot be violated unless stability is abandoned. There is also an important thermodynamic identity diQ≡diW ≥ 0 with important physical implications as it generalizes the well-known result of Count Rumford and the Gouy-Stodola theorem of classical thermodynamics. The μNEQT has far-reaching consequences with new results, and presents a new understanding of thermodynamics even of an isolated system at the microstate level, which has been an unsolved problem. We end the review by applying it to three different problems of fundamental interest.