Deep Process Research Articles

Influence of deep rolling at different temperatures on near-surface and mechanical properties of a Maraging C250 steel

Due to an outstanding combination of high strength and fracture toughness Maraging steels are used in numerous industries, e.g., in the aerospace sector. The high strength can be attributed to precipitation strengthening by conventional age-hardening heat treatments. However, as many of the envisaged applications will suffer from cyclic loadings, the fatigue properties are of significant interest in terms of a reliable use. Generally, Maraging steels are characterized by a relatively low fatigue strength limiting the use in certain applications. In this context, mechanical surface treatment processes are promising candidates to overcome these challenges. In particular, deep rolling represents a suitable process, as it is known for increased depth effects as well as smoother surface profiles compared to other mechanical surface treatment processes. Moreover, aging heat treatments can easily be implemented into the process. Accordingly, the present study focuses on near-surface and mechanical properties of a Maraging C250 steel after deep rolling. Characterization of the near-surface area was conducted by X-ray diffraction and hardness measurements. Mechanical properties were assessed by tensile and fatigue tests. The results presented reveal a great potential of the integration of an aging heat treatment into the deep rolling process eventually leading to significant process time reduction.

Getting better? Examining the effects of social support in OHCs on users’ emotional improvement

Online health communities (OHCs) have emerged as dynamic platforms facilitating patient communication and knowledge exchange on health matters. While extant research underscores the emotional benefits accrued from engagement in OHCs, the precise quantification of how social support within these communities influences users’ emotional states remains an intricate challenge. In light of this, the primary objective of this study is to explore the influential mechanisms underlying the received social support on the emotional improvement of users in OHCs, while concurrently investigating the potential moderating effects posed by individuals’ health conditions and levels of community involvement. Leveraging authentic user interaction data harvested from an online diabetes community, this study employed deep learning models and natural language processing methods to gauge the magnitude of received social support and the variation of users’ emotional states. Subsequently, a multiple linear regression model was formulated to evaluate the effects of both informational and emotional support on users' emotional improvement. Empirical results reaffirm the pivotal role of both informational and emotional support in enhancing users’ emotional states, with emotional support emerging as particularly influential. Notably, prolonged disease duration amplifies the beneficent impact of social support on emotional enhancement. The research outcomes bear implications for ameliorating negative emotional experiences and effectively managing chronic diseases.

Core Technologies in Recommender Systems: Investigating and Analyzing Standard Implementations

This paper places special emphasis on the evolution of recommendation algorithms in the context of big data and machine learning, underscoring significant advancements in deep learning and natural language processing that have markedly improved the precision and personalization of recommendations. We explore case studies in various sectors, including e-commerce, streaming services, and social media, to illustrate the adaptation of these technologies to distinct industry requirements. A critical component of our analysis is the examination of the impact of user data privacy regulations on the design and functionality of these systems. Additionally, the paper addresses the challenges and prospective directions in the field, with a particular focus on ethical considerations, the mitigation of bias, and the incorporation of artificial intelligence for dynamic, context-aware recommendation systems. Our research methodology amalgamates an extensive literature review with empirical data analysis, offering an in-depth understanding of the current state of recommendation system technologies. The findings of this study are intended to contribute to the field by presenting a comprehensive view of the technological, ethical, and practical dimensions of recommendation systems in the contemporary digital landscape.

Research on the preparation and quality control process of cattle liquid tallow

Liquid butter is a special process with unique flavor, which has strong commercial value in the deep processing of animal fat. In this paper, the preparation of liquid butter was based on centrifugal supernatant as the main raw material, and the process of preparing liquid butter by ultrasonic-assisted transesterification method was used to optimize the preparation of liquid butter by single factor and orthogonal experimental design according to the sensory, freezing point and iodine value of the product. The experimental results showed that the optimal formula of liquid butter was as follows: liquid-to-material ratio of 1:1, ultrasonic temperature of 50°C, and ultrasonic time of 50min. Using this ratio, the sensory score of liquid butter prepared from centrifugal supernatant was 87 points, the freezing point was 22.9°C, and the iodine value was 67.5g/100g. The liquid butter prepared by this process is clear and translucent, with unique flavor, and has certain nutritional and health care effects, which has certain reference value for future application in the field of deep processing of animal fat.

Combined Analysis of Metabolomics and Biochemical Changes Reveals the Nutritional and Functional Characteristics of Red Palm Weevil Rhynchophus ferrugineus (Coleoptera: Curculionidae) Larvae at Different Developmental Stages.

In this study, the changes in the conventional nutrient and mineral compositions as well as the metabolomics characteristics of the red palm weevil (RPW) Rhynchophus ferrugineus Olivier (Curculionidae: Coleoptera) larvae at early (EL), middle (ML) and old (OL) developmental stages were investigated. Results showed that the EL and ML had the highest content of protein (53.87 g/100 g dw) and fat (67.95 g/100 g), respectively, and three kinds of RPW larvae were all found to be rich in unsaturated fatty acids (52.17-53.12%), potassium (5707.12-15,865.04 mg/kg) and phosphorus (2123.87-7728.31 mg/kg). In addition, their protein contained 17 amino acids with the largest proportion of glutamate. A total of 424 metabolites mainly including lipids and lipid-like molecules, organic acids and their derivatives, organic heterocycle compounds, alkaloids and their derivatives, etc. were identified in the RPW larvae. There was a significant enrichment in the ABC transport, citrate cycle (TCA cycle), aminoacyl-tRNA biosynthesis, and mTOR signaling pathways as the larvae grow according to the analysis results of the metabolic pathways of differential metabolites. The water extract of EL exhibited relatively higher hydroxyl, 2,2-diphenyl-1-pyrroline hydrochloride (DPPH) and 2,2'-azobis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical-scavenging ability with the EC50 values of 1.12 mg/mL, 11.23 mg/mL, and 2.52 mg/mL, respectively. These results contribute to a better understanding of the compositional changes of the RPW larvae during its life cycle and provide a theoretical grounding for its deep processing and high-value utilization.

Research Progress on Removal of Oxygenates in Fischer-Tropsch Oil by Adsorption

The oxygen-containing compounds in Fischer Tropsch synthetic oil greatly affect the downstream deep processing of hydrocarbons, and effective removal is required. Compared to traditional removal technologies such as hydrogenation deoxygenation, solvent extraction, and extraction distillation, adsorption deoxygenation technology has the advantages of low cost, mild operating conditions, easy removal and recovery, and minimal impact on oil quality. Therefore, adsorption deoxygenation technology has devel‐ oped rapidly in various removal processes and has become a research hotspot in the current Fischer Tropsch oil deoxygenation. Adsorbents are the core of adsorption deoxygenation technology. Therefore, this article briefly introduces the adsorption mechanism and summarizes the research progress of adsorbents widely used in recent years, such as silica gel, alumina, molecular sieves, and metal organic frameworks, in adsorbing oxygen-containing compounds in Fischer Tropsch synthetic oils. And provide reference suggestions for further adsorption and deoxygenation directions in the future.

Assessment of Gender Differences in Letters of Recommendation for Physical Therapy Residency Applications.

Letters of recommendation (LOR) are an integral component of physical therapy residency applications. Identifying the influence of applicant and writer gender in LOR will help identify whether potential implicit gender bias exists in physical therapy residency application processes. Several medical and surgical residency education programs have reported positive, neutral, or negative LOR female gender bias among applicants and writers. Little research exists on gender differences in LOR to physical therapy education programs or physical therapy residency programs. Seven hundred sixty-eight LOR were analyzed from 256 applications to 3 physical therapy residency programs (neurologic, orthopaedic, sports) at one institution from 2014 to 2020. Thematic categories were developed to identify themes in a sample of LOR. Associations between writer and applicant gender were analyzed using summary statistics, word counts, thematic and psycholinguistic extraction, and rule-based and deep learning Natural Language Processing . No significant difference in LOR word counts were found based on writer or applicant gender. Increased word counts were seen in sports residency LOR compared with the orthopaedic residency. Thematic analysis showed LOR gender differences with male applicants receiving more positive generalized recommendations and female applicants receiving more comments regarding interpersonal relationship skills. No thematic or psycholinguistic gender differences were seen by LOR writer. Male applicants were 1.9 times more likely to select all male LOR writers, whereas female applicants were 2.1 times more likely to choose all female LOR writers. Gender differences in LORs for physical therapy residencies were found using a comprehensive Natural Language Processing approach that identified both a positive recommendation male applicant gender bias and a positive interpersonal relationship skill female applicant gender bias. Applicants were not harmed nor helped by selecting LOR writers of the opposite gender. Admissions committees and LOR writers should be mindful of potential implicit gender biases in LOR submitted to physical therapy residency programs.

TRANSFORMING FINTECH FRAUD DETECTION WITH ADVANCED ARTIFICIAL INTELLIGENCE ALGORITHMS

The rapid evolution of financial technology (fintech) platforms has exponentially increased the volume and sophistication of financial transactions, concurrently elevating the risk and complexity of fraudulent activities. This necessitates a paradigm shift in fraud detection methodologies towards more agile, accurate, and predictive solutions. This paper presents a comprehensive study on the transformative potential of advanced Artificial Intelligence (AI) algorithms in enhancing fintech fraud detection mechanisms. By leveraging cutting-edge AI techniques including deep learning, machine learning, and natural language processing, this research aims to develop a robust fraud detection framework capable of identifying, analyzing, and preventing fraudulent transactions in real-time.
 Our methodology encompasses the deployment of several AI algorithms on extensive datasets comprising genuine and fraudulent financial transactions. Through a comparative analysis, we identify the most effective algorithms in terms of accuracy, efficiency, and scalability. Key findings reveal that deep learning models, particularly those employing neural networks, outperform traditional machine learning models in detecting complex and nuanced fraudulent activities. Furthermore, the integration of natural language processing enables the extraction and analysis of unstructured data, significantly enhancing the detection capabilities.
 Conclusively, this paper underscores the critical role of advanced AI algorithms in revolutionizing fintech fraud detection. It highlights the superior performance of AI-based models over conventional methods, offering fintech platforms a more dynamic and predictive approach to fraud prevention. This research not only contributes to the academic discourse on financial security but also provides practical insights for fintech companies striving to safeguard their operations against fraud.
 Keywords: Artificial Intelligence, Fintech, Fraud Detection, Ethical Ai, Regulatory Compliance, Data Privacy, Algorithmic Bias, Predictive Analytics, Blockchain Technology, Quantum Computing, Interdisciplinary Collaboration, Innovation, Transparency, Accountability, Continuous Learning, Ethical Principles, Real-Time Processing, Financial Sector.

Machine learning in cardiac stress test interpretation: a systematic review.

Coronary artery disease (CAD) is a leading health challenge worldwide. Exercise stress testing is a foundational non-invasive diagnostic tool. Nonetheless, its variable accuracy prompts the exploration of more reliable methods. Recent advancements in machine learning (ML), including deep learning and natural language processing, have shown potential in refining the interpretation of stress testing data. Adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we conducted a systematic review of ML applications in stress electrocardiogram (ECG) and stress echocardiography for CAD prognosis. Medical Literature Analysis and Retrieval System Online, Web of Science, and the Cochrane Library were used as databases. We analysed the ML models, outcomes, and performance metrics. Overall, seven relevant studies were identified. Machine-learning applications in stress ECGs resulted in sensitivity and specificity improvements. Some models achieved rates of above 96% in both metrics and reduced false positives by up to 21%. In stress echocardiography, ML models demonstrated an increase in diagnostic precision. Some models achieved specificity and sensitivity rates of up to 92.7 and 84.4%, respectively. Natural language processing applications enabled the categorization of stress echocardiography reports, with accuracy rates nearing 98%. Limitations include a small, retrospective study pool and the exclusion of nuclear stress testing, due to its well-documented status. This review indicates the potential of artificial intelligence applications in refining CAD stress testing assessment. Further development for real-world use is warranted.

Application of artificial intelligence digital holography technology based on medical sensors in the development of medical image fusion

With the rapid development of artificial intelligence and medical technology, sensor technology, as an important tool to obtain medical data, provides a wealth of physiological parameter information, but the traditional medical image fusion methods are difficult to make full use of the information provided by sensors. Therefore, this paper aims to explore the application of artificial intelligence digital holographic technology based on medical sensor in medical image fusion, and improve the accuracy and effect of medical image diagnosis and treatment. Based on the physiological parameter data obtained by medical sensor, combined with deep learning and image processing algorithm, a medical image fusion method based on artificial intelligence digital holography technology was proposed. The method uses the physiological parameter information provided by the sensor and combines multiple medical image sources to realize holographic medical image presentation. The experimental results show that the artificial intelligence digital holographic technology based on medical sensor can effectively improve the resolution and information content of medical image fusion. Compared with traditional medical image fusion methods, this method has higher image quality and more accurate diagnosis results.

Classification of different types of DDR RAM using transfer learning in convolutional learning networks

Technology, specifically computers play an important role in modern society. People who are new to computers can determine what type of RAM they have, which can be used to avoid confusion on what type of RAM their computer needs with the help of an Android device. For this study, a powerful computer with a Graphics Processing Unit (GPU) needed to be used to shorten the amount of time that the deep learning process takes. The study gathered images of 4 types of Random Access Memory for a RAM classification system. There were 1000 images in total for DDR1, DDR2, DDR3, and DDR4 RAM. The study utilized transfer learning to RAM type classification with pre-trained models such as VGG16, VGG19, Inception V3, and Xception. The data that was gathered showed that Xception is the best classifier with an initial average accuracy of 85.034% and a 100% Val_Accuracy even though the model had the longest loading time with 12 seconds.

AI Driven Applications for Precision Weed Management in Agricultural Crops

Agriculture plays a most important role in our Indian economy and therefore lowering the cost of production and improving the quality of agricultural products is highly demanded. A weed is a plant which grows in wrong place at the wrong time and doing mostly harm than crops. Weed competes with the crops for water, light, nutrients and space, and therefore it prevents crop yields. This paper proposes a new method in a contrary way, which combines deep learning and image processing technology to prevent these weeds. Machine learning technologies, are becoming crucial in agriculture to increase productivity, where advanced automation and control have been required. Based on large training datasets and pre-trained models, (Deep Learning) DL-based Convolutional Neural Networks (CNN) methods have proven to be more accurate than previous traditional techniques. Recently, Deep Learning (DL) has gained much attention due to its advantages in object detection, classification, and feature extraction. The system implementation of image processing technique for weed detection, a trained image is taken as a sample in order to demonstrate the difference between weed and the crop. Yolo frame work is used for annotate boundary boxes to the image with datasets. The effectiveness of the (You Only Live Once) YOLO-WEED system for real-time Unmanned Aerial Vehicle (UAV) weed detection, given its high speed and high accuracy in detection. After certain steps, we get desired output, where the weeds are separated from the crop that has been taken in the sample image. Key Words: DeepLearning,CNN, Unmanned Aerial Vehicle (UAV), Precision Weed Detection.

Quality detection and variety classification of pecan seeds using hyperspectral imaging technology combined with machine learning

Pecan (Carya illinoinensis K.), a well-known dried seed and woody oil tree, faces challenges in its industry due to complex quality assessment methods and confusing varieties. These challenges have seriously hampered the development of a large-scale pecan deep processing industry. This work aimed to apply hyperspectral imaging technology (HSI) combined with machine learning to evaluate the quality of pecan seeds and perform variety classification. The samples of this work were composed of 19 varieties of pecan seeds, with 30 seeds per variety. After spectral preprocessing, spectral features were extracted from the spectral profiles using feature extraction methods. Back-propagation neural network models and partial least squares models were established to predict the contents of crude fat and moisture in pecan seeds. Predictions of the best models gave good results with R2-score of 0.887 for the crude fat model and 0.950 for the moisture model. Additionally, support vector machine models were developed to identify pecan varieties. The model achieved good results in 19 pecan varieties identification with accuracy of 0.965. In conclusion, the combination of HSI and machine learning could be an effective tool in improving the pecan industry and providing sustainable and efficient methods in the production of pecan seeds.

Flexible and fine-grained simulation of speed in language processing.

According to the embodied cognition theory, language comprehension is achieved through mental simulation. This account is supported by a number of studies reporting action simulations during language comprehension. However, which details of sensory-motor experience are included in these simulations is still controversial. Here, three experiments were carried out to examine the simulation of speed in action language comprehension. Experiment 1 adopted a lexical decision task and a semantic similarity judgment task on isolated fast and slow action verbs. It has been shown that fast action verbs were processed significantly faster than slow action verbs when deep semantic processing is required. Experiment 2 and Experiment 3 investigated the contextual influence on the simulation of speed, showing that the processing of verbs, either depicting fast actions or neutral actions, would be slowed down when embedded in the slow action sentences. These experiments together demonstrate that the fine-gained information, speed, is an important part of action representation and can be simulated but may not in an automatic way. Moreover, the speed simulation is flexible and can be modulated by the context.

Combining Artificial Intelligence and Simplified Image Processing for the Automatic Detection of Mycobacterium tuberculosis in Acid-fast Stain : A Cross-institute Training and Validation Study.

Tuberculosis (TB) poses a significant health threat in Taiwan, necessitating efficient detection methods. Traditional screening for acid-fast positive bacilli in acid-fast stain is time-consuming and prone to human error due to staining artifacts. To address this, we present an automated TB detection platform leveraging deep learning and image processing. Whole slide images from 2 hospitals were collected and processed on a high-performance system. The system utilizes an image processing technique to highlight red, rod-like regions and a modified EfficientNet model for binary classification of TB-positive regions. Our approach achieves a 97% accuracy in tile-based TB image classification, with minimal loss during the image processing step. By setting a 0.99 threshold, false positives are significantly reduced, resulting in a 94% detection rate when assisting pathologists, compared with 68% without artificial intelligence assistance. Notably, our system efficiently identifies artifacts and contaminants, addressing challenges in digital slide interpretation. Cross-hospital validation demonstrates the system's adaptability. The proposed artificial intelligence-assisted pipeline improves both detection rates and time efficiency, making it a promising tool for routine pathology work in TB detection.

Automated Pavement Condition Index Assessment with Deep Learning and Image Analysis: An End-to-End Approach.

The degradation of road pavements due to environmental factors is a pressing issue in infrastructure maintenance, necessitating precise identification of pavement distresses. The pavement condition index (PCI) serves as a critical metric for evaluating pavement conditions, essential for effective budget allocation and performance tracking. Traditional manual PCI assessment methods are limited by labor intensity, subjectivity, and susceptibility to human error. Addressing these challenges, this paper presents a novel, end-to-end automated method for PCI calculation, integrating deep learning and image processing technologies. The first stage employs a deep learning algorithm for accurate detection of pavement cracks, followed by the application of a segmentation-based skeleton algorithm in image processing to estimate crack width precisely. This integrated approach enhances the assessment process, providing a more comprehensive evaluation of pavement integrity. The validation results demonstrate a 95% accuracy in crack detection and 90% accuracy in crack width estimation. Leveraging these results, the automated PCI rating is achieved, aligned with standards, showcasing significant improvements in the efficiency and reliability of PCI evaluations. This method offers advancements in pavement maintenance strategies and potential applications in broader road infrastructure management.

Monkeypox Detection using CSA Based K-Means Clustering with Swin Transformer Model

Despite the global COVID-19 pandemic, public health professionals are also concerned about a possible new monkeypox epidemic. Similar to vaccinia, cowpox, and variola, the orthopoxvirus that causes monkeypox has two strands that are double-stranded. Many people have propagated the current pandemic through sexual means, particularly those who identify as bisexual or gay. The speed with which monkeypox was detected is the most important element here. In order to catch monkeypox before it infects more people, machine learning could be a huge help in making a quick and accurate diagnosis. Finding a solution is the driving force behind this project, which aims to develop a model for detecting monkeypox using deep learning and image processing. For optimal cluster selection during photo segmentation, the Chameleon Swarm Algorithm (CSA) employs K-means clustering. Examining the accuracy with which the Swin Transformer model identified instances of monkeypox was the driving force for this study. The proposed techniques are evaluated on two datasets: Kaggle Monkeypox Skin Lesion Dataset (MSLD) besides the Monkeypox Skin Image Dataset (MSID). We assessed the outcomes of various deep learning models using sensitivity, specificity, and balanced accuracy. Positive results from the projected process raise the possibility of its widespread application in monkeypox detection. This ingenious and cheap method can be put to good use in economically deprived communities that may not have access to proper laboratory facilities.

Allanite U–Pb dating places new constraints on the high‐pressure to high‐temperature evolution of the deep Himalayan crust

AbstractDuring continental collision, crustal rocks are buried, deformed, transformed and exhumed. The rates, timescales and tectonic implications of these processes are constrained through the sequence and conditions of metamorphic reactions in major and accessory phases. Petrographic, isotopic and elemental data from metabasite samples in NW Bhutan, eastern Himalaya, suggest initial equilibration under high‐pressure (plagioclase‐absent and rutile‐present) conditions, followed by decompression to lower pressure conditions at high‐temperatures that stabilized plagioclase, orthopyroxene and ilmenite. Field observations and chemical indicators suggest equilibration under the lower pressure conditions is likely linked to the infiltration of melt from the host metasedimentary rocks. The metabasites preserve two metamorphic growth stages of chemically‐and petrographically distinct allanite that temporally overlap two stages of zircon growth. Allanite cores and zircon mantles grew at c. 19 ± 2 and 17–15.5 Ma respectively, linked texturally and chemically to the high‐pressure evolution. Symplectitic rims on embayed allanite cores, wholly symplectized Aln–Ilm and Aln–Cpx grains, and high U zircon rims grew at c. 15.5–14.5 Ma, linked chemically to the presence of melt and lower pressure, high‐temperature conditions. A single garnet Lu–Hf date is interpreted as geologically meaningless, with the bulk rock composition modified by melt infiltration after garnet formation. The open system evolution of these rocks precludes precise determination of the reactive bulk composition during metamorphic evolution and thus absolute conditions, especially during the early high‐pressure evolution. Despite these limitations, we show that combined geochemical and petrographic datasets are still able to provide insights into the rates and timescales of deep orogenic processes. The data suggest a younger and shallower evolution for the NW Bhutan metabasites compared to similar rocks in the central and eastern Himalayas.

Deep Learning and Natural Language Processing Technology Based Display and Analysis of Modern Artwork

The modern artwork analysis display system, empowered by natural language processing (NLP) technology, revolutionizes the way audiences interact with and understand art. By integrating NLP algorithms, this system offers a dynamic and user-friendly platform for analyzing and displaying artwork. Utilizing NLP, visitors can engage in interactive conversations with the system, asking questions or making inquiries about the artwork on display. The system processes these inquiries, extracting relevant information from curated databases and scholarly sources to provide insightful and context-rich responses. Additionally, NLP algorithms can analyze textual descriptions, artist statements, and critical reviews to offer nuanced interpretations and historical context for each artwork. This paper presents the design and implementation of an innovative modern artwork analysis and display system, leveraging deep learning and natural language processing (NLP) technology, integrated with Multi-Feature Extraction Fuzzy Classification (MFEFC). The system offers a comprehensive platform for analyzing and presenting modern artworks, enhancing user engagement and understanding. Deep learning algorithms are employed to extract high-level features from visual artworks, allowing for automatic recognition of artistic styles, genres, and themes. Concurrently, NLP techniques process textual descriptions, artist biographies, and critical reviews to provide contextual information and interpretative insights. The integration of MFEFC enables precise classification of artworks based on multiple features extracted from both visual and textual sources, facilitating accurate analysis and categorization. Simulation of the NLP techniques demonstrated an average precision of 90% in extracting relevant contextual information from textual descriptions and artist biographies. Furthermore, MFEFC achieved a classification accuracy of 88% in categorizing artworks based on combined visual and textual features.

Seismic structure of the 2015 Mw7.8 Gorkha earthquake revealed by ambient seismic noise and teleseismic surface wave tomography

The destructive 2015 Mw7.8 Gorkha earthquake occurred in the Main Himalayan Thrust due to the collision of the Indian and Asian plates, which provides a unique opportunity to understand the deep dynamic processes and seismogenic mechanisms of strong earthquakes. We construct a regional-scale shear-wave velocity model of the crust and uppermost mantle using ambient seismic noise and teleseismic surface wave at periods of 5–100 s around the Gorkha earthquake region. The new shear-wave velocity model exhibits prominently lateral heterogeneities in the Gorkha earthquake areas. We observe a high-velocity (high-V) zone around the Gorkha main shock in the Main Himalayan Thrust, indicating the existence of a high-strength asperity that sustains the stress accumulating. The aftershocks are primarily located in the low-velocity (low-V) anomalies and enclosed by two high-V anomalies, which appear to act as structural barriers that influence the spread of the aftershocks. Prominent low-Vanomalies from the lower crust to the mantle lithosphere are observed along the north–south trending rifts, suggesting the hot materials upwelling due to the tearing of the northward subducting Indian lithosphere. These observations may indicate that seismic velocity heterogeneity could play an essential role in earthquake initiation and the rupture process.