Standard Algorithm Research Articles

A Complete Study of Remote Sensing- Sentinel-2 Satellite Data for Land Use / Land Cover (LULC) Analysis

Remote sensing based Satellite collected Images is a complete advanced process of both the automatic detecting observation and overall earth monitoring as well as observing the mainly physical level of characteristics of an area by mainly measuring its reflected and emitted radiation at a distance mainly used from both satellite and aircraft. Special quality cameras collect remotely based sensed images, which help all researchers "sense" things about overall Earth observation. The European Research Space Agency (ESA) and the European (member countries) Union-EU both have equally provided together to specially a very powerful observation of the whole planet surface by simply making the remote sensing based Sentinel type-2 numerous spectral based results. The Sentinel category 2 satellite was successfully launched on June, 23rd 2015 to deliver satellite collection of data and imagery. The sentinel category-1 is the mission of the introductory satellite of Copernicus oriented research project for the artificial satellite based asterism organized by all Europe countries Space Research Agency. The satellite series of Sentinel 1A was successfully produced in the month of April, 03 2014, the same as Sentinel-1B satellite was successfully launched in the month of April, 25 2016. The Sentinel category-2 is the second category of satellite constellation of the All European countries Space Research Program designed mainly for a series of Sentinel missions. The main goal of the Sentinel category-2 satellite series mission is primary provide a very good quality of best pixel-resolution space satellite data / information used mainly for Land surface Cover (LC)/ Land Usage (LU) observation, coastal surface observation, different climate (Cloud) change and natural disaster damage mapping, Natural disaster, Earth Observatory, Forestry, Wildfire Assessment, Humanitarian operations with a best complementing to the other remote sensing satellite series missions mainly as-”LANDSAT”. Therefore, the review of advanced Sentinel type -2 is primary highlighted on these two important parameters: First, The main advantage of the ESA-Sentinel type 2 to different Land surface cover / Land usage classification or monitoring, second, Discovering total overall results of the ESA- Sentinel type-2 satellite data / imagery in various remote sensing based applications (such as the green-forest monitoring, urban area locality, natural level incident hazard analyze monitoring, coastal monitoring). The complete overall accuracy of data analysis can be improved when the possibility to simply integrate Sentinel type-2 satellite data with additional various types of advanced remotely based satellite observed details. The overall literature review of the Sentinel type-2 also describes a use of advanced remote sensing Sentinel type-2 details which are produced a very high level accuracies (>81 to 85% by using various advanced machine-learning classification algorithms mainly as: Logistic standard Regression, Support based-Vector (classification) Machine-SVM, Decision level of Trees and Random standard forest-RF algorithm. Remaining, other classifier techniques commonly known as Maximum-Likelihood Analysis (MLA) are most commonly used. Although the sentinel type-2 also provides good level of opportunities mainly for Land surface cover-LC / Land usage-LU based on various classification, as well as they have many complex difficulties which added different mismatching with data of Landsat satellite based on OLI-category 8 additionally a gap of various thermal level bands, and different spatial based resolution (pixel) among the various bands of Sentinel type-2 satellite. The Sentinel-2 data has the good quality of probable to significantly present in the direction of land surface coverage with land usage monitoring as well as classification.

The Challenges of Patients with Spinal CSF Leaks in Canada: A Cross-Sectional Online Survey.

Spinal CSF leak can cause disabling headaches and neurological symptoms. Lack of awareness, diagnostic delay and treatment inconsistencies affect the quality of CSF leak care globally. This is the first study aiming to identify and assess these challenges in Canada. A cross-sectional online survey of Canadian patients with spinal CSF leak was designed in collaboration with Spinal CSF Leak Canada, including questions on demographics, headache condition, investigations, treatments, quality of life, financial consequences and out-of-country care. The survey captured 103 respondents with confirmed spinal CSF leak diagnosis, of whom 56% were still suffering. The majority were female (80%), most being highly educated, with a mean age of 41.8 (SD: 10.37) years at the time of diagnosis. Inconsistencies in care resulted in variable durations for obtaining diagnosis and treatment. The majority of respondents (88%) had seen multiple physicians, and only 50% had seen a CSF leak specialist. Invasive imaging was not performed in 43%. CSF leak relapse after initial successful treatment occurred frequently (43%). The incidence of rebound intracranial hypertension was high (52.5%), and the treatment was difficult to access (77%). Out-of-country care was common (28%), and the impact on financial health was omnipresent (81.5%). The survey demonstrates significant gaps in spinal CSF leak care in Canada, similar to global observations. Lack of awareness and access, delayed care, and inconsistencies in investigations and management are common. Spinal CSF leak significantly impacts patients' physical, mental and financial well-being. Increased awareness, referral pathways and standardized treatment algorithms are key factors in optimizing patient care in Canada.

EGS SO32 - Management of Spontaneous Oesophageal Perforations in the era of Endo-Vacuum Therapy- Preliminary Results of a Single-Institution Study

Abstract Background Spontaneous Oesophageal Perforation (SOP) is a rare gastro-intestinal surgical emergency with devastating morbidity and mortality. Given the rarity of the entity, there is no up-to-date evidence on outcomes in the context of centralized esophagogastric services neither consensus guidelines nor standardized treatment algorithms on how these should be optimally managed. Endoscopic Vacuum Therapy (EVT) has emerged as safe and effective option particularly in cases where major invasive surgery would otherwise be deferred. Herein, we sought to explore the effectiveness, patient benefit and related morbidity/mortality by exploring the strategy outcomes in SOPs prior to and during periods where EVT was available. Method A single-centre retrospective cohort study incorporating all consecutive patients (N=72) presented or transferred to our tertiary oesophagogastric unit for SOP management between 01/01/2009-31/12/2023 was undertaken. Patients with anastomotic leak post-surgery or perforation secondary to malignancy, iatrogenic, traumatic or food bolus were excluded. Primary study-outcomes were treatment type including non-operative, EVT after 01/01/2018 or other endoscopic therapy, surgical operation including laparotomy, thoracotomy, primary defect repair or over a T-tube, disconnection with oesophagostomy, stenting or other interventional radiology procedures and their rate of success or failure. Secondary outcomes were length of stay, 90-day complications requiring re-intervention or re-admission, 30-, 90-day, 1-year mortality. Results In 2009-2017 pre-EVT period, 25/72 patients were managed a) 18/25 surgically with thoracotomy and T-tube, b) non-surgical 7/25, four were palliated. 30-day, 90-day mortality were 4/25 and 2/25, 90-day morbidity Clavien-Dindo >IIIa 14/25 (56%), 90-day reintervention 8/25 (32%), median LOS 27 days. In 2018-2023 EVT period , 47/72 patients were treated with a) laparotomy, left thoracotomy, defect repair over t-tube, chest drainage in 18/47, b) non-surgical 15/47, c) endoscopy 6/47, d) EVT in 14/47. 30-day mortality was 6/47, 90-day mortality 2/47, 90-day morbidity Clavien-Dindo >IIIa 32/47 (68%), 90-day reintervention 26/47 (55.3%), median LOS 34 days. 1-year mortality was 5/25 and 3/47 respectively. Conclusion EVT therapy appears to be a feasible additional treatment modality in the management of spontaneous oesophageal perforations. Optimal patient selection, therapeutic endoscopic expertise, knowledge of the technique, multidisciplinary approach and team collaboration are essentials for favourable outcomes. Of note, management of oesophageal perforations in a dedicated UGI centre is directly related to patient outcomes and is recommended as soon as such a diagnosis is confirmed.

Cataract surgery combined with glaucoma surgery.

This review evaluates the surgical management of glaucoma in patients undergoing cataract surgery. Combining both procedures present challenges in balancing intraocular pressure (IOP) control, minimizing postoperative complications, and achieving optimal refractive outcomes. Recent studies highlight the effectiveness of combined glaucoma and cataract surgeries, with traditional filtering surgeries and MIGS showing significant IOP reduction. The abundance of long-term studies shows that MIGS can offer an effective and safer alternative when carefully tailored to meet the specific needs of each patient. Combining glaucoma and cataract surgery is a promising approach for patients with coexisting conditions. While traditional surgeries offer robust IOP reduction, MIGS procedures offer better safety profiles with fewer complications and more predictable refractive results. Surgeons must carefully consider the timing and choice of procedures, with further research required to develop standardized treatment algorithms.

Surgical Management of Symptomatic Recurrent/Progressive Craniopharyngioma: Is There a Gold-Standard Approach?

The standard treatment for craniopharyngiomas (CPs) involves either initial gross total resection or subtotal resection with adjuvant radiotherapy. However, there is no consensus regarding the management of recurrent cases. We reviewed a series of patients with CP to evaluate the characteristics of patients with recurrent/progressing CP. We retrospectively reviewed 31 patients with biopsy-proven recurrent/progressed CP who were initially operated on at our clinic from 2015 to 2023. Demographic and clinical data, including age, sex, tumor localization, lesion size and features, radiological findings, and endocrine, neurologic, and ophthalmological issues, were compared between those with and without recurrence. Fifteen out of 31 patients (48.4%) underwent transnasal/transsphenoidal endoscopic surgery; the remaining 16 underwent microscopic transcranial or neuroendoscopic surgery. After treatment, recurrence and events (any medical condition related to the patient's current condition other than recurrence) were observed in 48.4% and 32.3% of the cases, respectively. There were no statistically significant disparities between patients with and without recurrence in terms of age, tumor dimensions, tumor localization, gender distribution, surgical modality, or the presence of calcifications or hydrocephalus (p > 0.05). The highest and lowest overall survival was 25 years and 6 months, respectively. The management of recurrent CP is too complicated to fit into a standard treatment algorithm. Therefore, surgical treatment for these patients should be individualized and planned according to patient symptoms and radiological findings, which is presumably a more versatile, efficient, and safe approach for recurrent CPs.

Implementing the Grover algorithm in homomorphic encryption schemes

We apply quantum homomorphic encryption (QHE) schemes suitable for circuits with a polynomial number of T+T† gates to Grover's algorithm, performing a simulation in Qiskit of a Grover circuit that contains three qubits. The T+T†-gate complexity of Grover's algorithm is also analyzed in order to show that any Grover circuit can be evaluated homomorphically in an efficient manner. We discuss how to apply these QHE schemes to allow for the efficient homomorphic evaluation of any Grover circuit composed of n qubits using n−2 extra ancilla qubits. We also show how the homomorphic evaluation of the special case where there is only one marked item can be implemented using an algorithm that makes the decryption process more efficient compared with the standard Grover algorithm. Published by the American Physical Society 2024

Approaches to Pediatric Chest Pain: A Narrative Review

Chest pain in children and adolescents is a common reason for emergency department visits and referrals to pediatric cardiologists, often driven by parental concern about potential cardiac causes. However, the vast majority of pediatric chest pain cases are benign and non-cardiac in origin. This narrative review examines the etiology, evaluation, and management of pediatric chest pain, emphasizing the importance of a thorough clinical history and physical examination in distinguishing between benign and serious conditions. This review also explores the role of diagnostic tests such as electrocardiograms, chest radiography, and echocardiography, highlighting the need to balance the avoidance of unnecessary tests with the imperative to rule out life-threatening cardiac conditions. Despite the low prevalence of cardiac causes, the variability in diagnostic approaches underscores the need for standardized evaluation algorithms. These could streamline care, reduce unnecessary resource utilization, and minimize anxiety for both patients and their families. Future studies should focus on assessing the effectiveness of such algorithms in improving clinical outcomes and resource management. The findings underscore the importance of a careful, evidence-based approach to the management of pediatric chest pain.

Optimization of a wideband discrete Raman amplifier in a P2O5-doped optical fiber using multi-objective grey wolf algorithm

This paper demonstrates the use of the multi-objective grey wolf algorithm to optimize a discrete Raman amplifier (DRA) in a P2O5-doped optical fiber. Specifically, the multi-objective grey wolf algorithm is combined with the DRA to carry out a process that seeks to maximize the gain and minimize the ripple. The P2O5-doped optical fiber employed in this study has a Raman gain coefficient spectrum with multiple peaks with different frequency shifts. This allows them to be combined in more complex ways than optical fibers with a single peak in the Raman gain spectrum. Consequently, the gain curve produced with this fiber has the potential to be more adjustable even when fewer pumps are used. Thus, this paper explores this fact to perform, to the best of our knowledge, the first specialized optimization process reported in the scientific literature of a wideband discrete Raman amplifier in a P2O5-doped optical fiber. With a different gain profile of this fiber compared to those of traditional standard optical fibers, it was possible to design a wideband DRA, going from 1530 nm to 1675 nm, covering C+L+U bands, maintaining a ripple of up to 8 dB with a net gain of 14 dB using only 3 pumps. Moreover, this work demonstrates for the first time, through a comparative analysis, that the multi-objective grey wolf algorithm performs better than the standard and well-known non-dominated sorting genetic optimization algorithm to optimize a DRA in a P2O5-doped optical fiber. The proposed DRA is a feasible, low-cost, and simple alternative for building fiber amplifiers for future high-bandwidth and wideband wavelength division multiplexing (WDM) communication systems, network infrastructures such as data centers, undersea cables, 5G and beyond, and cutting-edge research applications.

Multi-objective optimization based robotic path planning for interpolation reconstruction model of CT data

Robotic path planning plays a pivotal role in computer-aided liver tumor thermal ablation surgery. However, traditional methods face challenges in terms of low reconstruction efficiency and planning safety. To address these issues, we propose a method for robotic path planning of liver tumor thermal ablation surgery. Firstly, an interlayer interpolation algorithm based on optical flow estimation is utilized to compensate for large interlayer spacing in computed tomography (CT) images by inserting predicted images between sequence images. Secondly, the voxel traversal strategy and patch intersection calculation strategy in the standard marching cube (MC) algorithm is optimized to improve the efficiency of abdominal tissues reconstruction. Finally, comprehensive clinical constraints are summarized to ensure the surgery safety and the strength pareto evolutionary algorithm II (SPEA-II) is leveraged to optimize surgery path, which can obtain even distribution of solutions in high-dimension optimization problems. Extensive experiments conducted on the 3Dircadb and SLIVER07 datasets revealed that our proposed method reduces reconstruction time by 21.5% compared to the standard MC algorithm, while achieving an average overlap rate of 88.25% and an average Hausdorff distance of 15.25 mm between Pareto front points and surgeon's recommended puncture points.

Parameter Estimation of Proton Exchange Membrane Fuel Cells Using Chaotic Newton-Raphson-Based Optimizer

This paper presents a novel improved metaheuristic algorithm, Chaotic Newton-Raphson-Based Optimizer (CNRBO), for the Proton Exchange Membrane Fuel Cells (PEMFCs) parameter extraction problem. In this study, ten distinct chaotic maps are integrated within the Newton-Raphson-based optimizer (NRBO) to improve its performance. The chaotic maps are employed to define the probability of the trap avoidance operator (TAO). The proposed CNRBO algorithm is validated using standard benchmark optimization problems. Results proved its advantages over the standard NRBO algorithm in terms of accuracy, robustness, and convergence speed. For PEMFC parameters extraction using the proposed CNRBO algorithm, the objective function to be minimized is the sum of squared errors (SSE) between estimated and measured stack voltages across various data points. To validate the effectiveness and reliability of the proposed CNRBO, its performance is compared with recent algorithms on eight different PEMFC stack models including NedStackPS6, BCS 500W, Horizon 500W, 250W stack, Avista SR-12 500W, Temasek 1KW, Ballard Mark V 5KW and Horizon H-1000XP stack. Statistical measures are employed to assess the superiority and robustness of the proposed CNRBO. Statistical analyses demonstrate that the proposed CNRBO algorithm outperforms existing algorithms in terms of accuracy, search capability, and convergence speed, solidifying its position as a powerful tool for PEMFC parameter estimation.

Using Multiple Drug Similarity Networks to Promote Adverse Drug Event Detection

The occurrence of an adverse drug event (ADE) has become a serious social concern of public health. Early detection of ADEs can lower the risk of drug safety as well as the expense of the drug. While post-market spontaneous reports of ADEs remain a cornerstone of pharmacovigilance, most existing signal detection algorithms rely on substantial accumulated data, limiting their applicability to early ADE detection when reports are scarce. To address this issue, we propose a label propagation model for generating enhanced drug safety signals using multiple drug features. We first construct multiple drug similarity networks using a range of drug features. We then calculate initial drug safety signals using conventional signal detection algorithms. These original signals are subsequently propagated across each drug similarity network to obtain enhanced drug safety signals. We evaluate our proposed model using two common signal detection algorithms on data from the FDA Adverse Event Reporting System (FAERS). Results demonstrate that enhanced drug safety signals with pre-clinical information outperform the standard safety signal detection algorithms on early ADE detection. In addition, we systematically evaluate the performance of different drug similarities against different types of ADEs. Furthermore, we have developed a web interface (http://drug-drug-sim.aimedlab.net/) to display our multiple drug similarity scores, facilitating access to this valuable resource for drug safety monitoring.

Generative Models for Grid-Based and Image-Based Pathfinding

Pathfinding is a challenging problem which generally asks to find a sequence of valid moves for an agent provided with a representation of the environment, i.e. a map, in which it operates. In this work, we consider pathfinding on binary grids and on image representations of the digital elevation models. In the former case, the transition costs are known, while in latter scenario, they are not. A widespread method to solve the first problem is to utilize a search algorithm that systematically explores the search space to obtain a solution. Ideally, the search should also be complemented with an informative heuristic to focus on the goal and prune the unpromising regions of the search space, thus decreasing the number of search iterations. Unfortunately, the widespread heuristic functions for grid-based pathfinding, such as Manhattan distance or Chebyshev distance, do not take the obstacles into account and in obstacle-rich environments demonstrate inefficient performance. As for pathfinding with image inputs, the heuristic search cannot be applied straightforwardly as the transition costs, i.e. the costs of moving from one pixel to the other, are not known. To tackle both challenges, we suggest utilizing modern deep neural networks to infer the instance-dependent heuristic functions at the pre-processing step and further use them for pathfinding with standard heuristic search algorithms. The principal heuristic function that we suggest learning is the path probability, which indicates how likely the grid cell (pixel) is lying on the shortest path (for binary grids with known transition costs, we also suggest another variant of the heuristic function that can speed up the search). Learning is performed in a supervised fashion (while we have also explored the possibilities of end-to-end learning that includes a planner in the learning pipeline). At the test time, path probability is used as the secondary heuristic for the Focal Search, a specific heuristic search algorithm that provides the theoretical guarantees on the cost bound of the resultant solution. Empirically, we show that the suggested approach significantly outperforms state-of-the-art competitors in a variety of different tasks (including out-of-the distribution instances).

Diagnosis and management of infections related to spinal pain interventions.

Although relatively rare, infections related to spinal pain interventions pose significant risks with an increase in the frequency of these procedures. This review investigates the incidence, risk factors, and management of infections following spinal pain interventions, such as epidural steroid injections, nucleoplasty, and facet joint injections. Most of the existing literature comprises case reports and retrospective studies with limited prospective research, owing to the nature of these infections. Our analysis revealed that while the overall infection rate is low, potential complications, such as epidural abscesses and spondylodiscitis, can be severe and life-threatening. The risk factors include advanced age, diabetes, immunosuppression, and multiple spinal procedures. Early diagnosis and timely intervention are critical to prevent long-term morbidity. These findings emphasize the importance of developing standardized diagnostic algorithms and treatment guidelines to support clinicians in managing these infections effectively. Future research should focus on large-scale studies to understand the impact of these infections better and refine clinical management strategies.

Temperature modeling of wave rotor refrigeration process based on variable time-delay compensation and SDAE-GRU

Abstract Aiming at the problem of noise and time-delay in temperature prediction modeling of large wave rotor refrigeration process, this paper proposes a novel modeling algorithm that combines the advantages of stacked denoising autoencoder (SDAE) network and Gate Recurrent Unit (GRU) algorithm and utilizes the Maximal Information Coefficient (MIC) algorithm to solve the targeting problem. The MIC algorithm is utilized to calculate the delay time of each input variable and output variable, and the input matrix is reconstructed to compensate for the time-delay, thus solving the problem afterwards. The SDAE network is utilized for denoising the input data and feature extraction to solve the problem of presence of noise in the data. The simulation results show that the proposed MIC-SDAE-GRU algorithm achieves a Root Mean Squared Error (RMSE) of 0.6432 and an R-squared (R²) value of 0.9638, outperforming traditional machine learning methods and other deep learning approaches. Specifically, compared to the standard GRU algorithm, it improves the accuracy of temperature prediction for the wave rotor refrigeration process by 24.7% and exhibits strong generalization ability across various operating conditions.

Data protection enhancement in smart grid communication: An efficient multi-layer encrypting approach based on chaotic techniques and steganography

In this study, a multi-layer encryption system is proposed through which sensitive and important data in smart grids can be secured from unauthorized persons as well as cyber threats. The suggested system is analyzed using fundamental performance assessment metrics say as average processing times across ten runs, peak signal-to-noise ratio (PSNR), and root mean square error (RMSE). The encryption process involves hiding the data encrypted by a digital dictionary and the advanced encryption standard (AES) algorithm in an image. In this way, AES encryption suggests rapid processing with a short average time per each block; similarly, the decryption technique, which includes the dictionary and AES decryption also renders the results in a justified computational time nominating it as a practical approach. It is worth noting that the small key era and dictionary construction instances provoke innovations in the general efficiency of the encryption mechanism. The conducted study, based on recorded values for PSNR, RMSE, UACI, and NPCR, demonstrates that the proposed multi-layer encryption system, which incorporates dictionary encryption and AES, is a robust and a fast method to shield important information and attaining a balance between protection and computational time.

Transarterial Embolization for Musculoskeletal Pain Management: AJR Expert Panel Narrative Review.

Musculoskeletal embolization has emerged in recent years as a treatment for chronic joint pain, as the inflammatory cascade responsible for such pain has become better understood. Studies have demonstrated a complex interplay between joint inflammation and synovial hypervascularity that causes growth of new unmyelinated nerve fibers responsible for pain. Embolization targets joint hypervascularity, to disrupt the inflammatory cycle and provide pain relief. The standard treatment algorithm for chronic joint pain is well-established and entails escalating therapeutic options that include exercise, self-management programs, analgesic medications, intra-articular injections, and finally surgical replacement or release. Genicular artery embolization (GAE), targeting abnormal vasculature around the knee joint, is the most heavily studied musucloskeletal embolization procedure, reflecting the high worldwide prevalence and increasing incidence of knee osteoarthritis. GAE is now supported by multiple prospective studies, including randomized control trials comparing GAE versus sham treatment. Embolization has also extended outside of the knee joint to include the shoulder (treatment of adhesive capsulitis or secondary stiff shoulder), elbow (medial/lateral epicondylitis), hip (osteoarthritis, great trochanteric pain syndrome), and ankle (plantar fasciitis). This AJR Expert Panel Narrative Review discusses the current status of transarterial embolization for musculoskeletal pain management, focusing on treatment of knee osteoarthritis and chronic shoulder pain.

Comparison of 2 Sets of Immunoassays Used in Modified 2-Tiered Testing Algorithms for the Diagnosis of Lyme Disease.

Since 2019, modified 2-tiered testing (MTTT) algorithms have been available for the diagnosis of Lyme disease. MTTTs replaced the standard algorithms that utilized enzyme immunoassays and immunoblots with sequential enzyme immunoassays that detect different antigens. We compared the performance of serological assays from ZEUS Scientific Inc. and DiaSorin Inc. that are used for the diagnosis of Lyme disease. Serological results were compared with clinical information gathered by chart review. Percent positive agreement (PPA) and percent negative agreement (PNA) for total immunoglobulin G (IgG)/immunoglogulin M (IgM) (n = 120) were 64% (95% confidence interval 54% to 73%) and 100% (87% to 100%), respectively. PPA and PNA for IgG (n = 93) were 91% (80% to 97%) and 66% (52% to 78%), respectively. PPA and PNA for IgM (n = 93) were 75% (62% to 85%) and 95% (82% to 99%), respectively. Fewer positive total IgG/IgM results confirmed positive for either IgG or IgM for ZEUS compared to DiaSorin. Overall MTTT algorithm interpretation was concordant in 58% (55/95) of samples, and concordance improved when the results were limited to IgM in patients with symptom duration <30 days. Treatment with antibiotics was most strongly associated with IgM positivity. This analysis highlights differences in the performance characteristics between commercially available diagnostic assays for Lyme disease. Our data suggest that the DiaSorin assays would result in fewer positive total IgG/IgM tests, decreasing the required number of confirmatory IgG and IgM tests. This would potentially lead to fewer patients treated with antibiotics.

The Advantage of Board Game with Deep Reinforcement Learning and Causal Inference

With the continuous progress of artificial intelligence technology, the deep reinforcement learning method combining behaviourism and connectionism has shown fantastic performance far beyond the human level in chess games. This paper systematically introduces two mainstream chess games, including the standard algorithms of artificial intelligence technology in chess and Go, including Monte Carlo and deep reinforcement learning, and expounds on their algorithm principles and the core reasons for their strong learning ability. In addition, this paper discusses the limitations of existing methods in interpretability and puts forward the possibility of applying causal reasoning to deep reinforcement learning to solve interpretable problems. This study will provide a valuable reference for practitioners in related fields.

State of Health Estimation of Lithium-Ion Batteries Using Fusion Health Indicator by PSO-ELM Model

The accurate estimation of the State of Health (SOH) of lithium-ion batteries is essential for ensuring their safe and reliable operation, as direct measurement is not feasible. This paper presents a novel SOH estimation method that integrates Particle Swarm Optimization (PSO) with an Extreme Learning Machine (ELM) to improve prediction accuracy. Health Indicators (HIs) are first extracted from the battery’s charging curve, and correlation analysis is conducted on seven indirect HIs using Pearson and Spearman coefficients. To reduce dimensionality and eliminate redundancy, Principal Component Analysis (PCA) is applied, with the principal component contributing over 94% used as a fusion HI to represent battery capacity degradation. PSO is then employed to optimize the weights (ε) between the input and hidden layers, as well as the hidden layer bias (u) in the ELM, treating these parameters as particles in the PSO framework. This optimization enhances the ELM’s performance, addressing instability issues in the standard algorithm. The proposed PSO-ELM model demonstrates superior accuracy in SOH prediction compared with ELM and other methods. Experimental results show that the mean absolute error (MAE) is 0.0034, the mean absolute percentage error (MAPE) is 0.467%, and the root mean square error (RMSE) is 0.0043, providing a valuable reference for battery safety and reliability assessments.

Modified Sand Cat Swarm Optimization (MSCSO) Based Feature Selection for Classification of Diseases in Tomato Leaves

Food obtained from the plants is one of the most basic requirements for the survival of humanity. However, plant diseases produce low-quality products that adversely affect the country's economy. Machine learning based detection methods for such diseases can reduce these effects. However, these methods involve plant images and usually suffer the curse of dimensionality in terms of features. Features selection improves the performance of machine learning algorithms. Feature selection method aims to find optimal number of features for a machine learning algorithm such as classification, clustering and many more. This work proposed a Modified Sand Cat Swarm Optimization (MSCSO) algorithm based feature selection to find the optimal number of features. The algorithm performs well over other standard algorithm such as GA, PSO, and GWO. The performance of different classifiers are analysed for classification of diseases in tomato leaves based on features selected by MSCSO.