Update Algorithm Research Articles

DCAGS-IoT: Dynamic Cross-Domain Authentication Scheme Using Group Signature in IoT

Cross-domain authentication requires that there is no trust gap between different trust domains that can cause cross-domain devices to exceed the security control scope of the original trust domain and further expose cross-domain authentication systems to security threats. In addition, as relying on the traditional cross-domain authentication means built by centralized institutions cannot meet the data security needs in a big data environment. Therefore, it is necessary to design a secure dynamic cross-domain authentication scheme. In this paper, we propose a dynamic cross-domain authentication scheme (DCAGS-IoT) in the Internet of Things environment using the group signature technology and the distributed system architecture of blockchain. Specifically aiming at the problem of increasing and revoking users in dynamic cross-domain authentication, a user update algorithm with the complexity of O (logN) was designed to manage users in the trust domain. Moreover, we used the characteristics that group signature users can sign on behalf of a group to protect the users’ privacy and track suspicious users. Since the size of the signature generated by the scheme is independent of the number of group members N and only depends on the security parameters λ, the efficiency of the protocol implementation is improved, and the security and availability of the authentication scheme are guaranteed.

Symptomatic intracranial tumors in pregnancy: an updated management algorithm. Illustrative case.

Intracranial tumors are infrequently encountered during pregnancy, and their diagnosis and management require a multidisciplinary approach to ensure the best possible outcomes for the mother and fetus. The pathophysiology and manifestations of these tumors are influenced by hormonal changes, hemodynamic modifications, and alterations in immunological tolerance that occur during pregnancy. Despite the complexity of this condition, no standardized guidelines exist. This study aims to highlight the key points of this presentation, along with the discussion of a possible management algorithm. The authors report the case of a 35-year-old woman who presented during the third trimester of pregnancy with severe signs of increased intracranial pressure (ICP) due to a posterior cranial fossa mass. The decision was made to stabilize the patient by placing an external ventricular drain to temporize her increased ICPs until the baby could be safely delivered via cesarean section. The mass was resected via suboccipital craniectomy 1 week postpartum. In considering treatment modalities and their timing in patients presenting with intracranial tumors during pregnancy, each patient should be managed on the basis of an individual treatment algorithm. Symptoms, prognosis, and gestational age should be taken into account to optimize the surgical and perioperative outcomes of both the mother and fetus.

Treatment of motor symptoms in Parkinson's disease.

In recent years, the development of new therapies and improvements in our understanding of older therapies have led to changes in the management of Parkinson's disease. However, current Norwegian and international therapy recommendations present a range of different options as being equally viable. In this clinical review, we propose an updated algorithm for the medical treatment of motor symptoms in Parkinson's disease, based on evidence-based recommendations and our own personal experience and opinions.

Statement of the Spanish Interdisciplinary Vascular Prevention Committee on the updated European Guidelines on Cardiovascular Disease Prevention

We present the Spanish adaptation of the 2021 European Guidelines on Cardiovascular Disease (CVD) prevention in clinical practice. The current guidelines besides the individual approach greatly emphasize on the importance of population level approaches to the prevention of cardiovascular diseases. Systematic global CVD risk assessment is recommended in individuals with any major vascular risk factor. Regarding LDL-Cholesterol, blood pressure, and glycemic control in patients with diabetes mellitus, goals and targets remain as recommended in previous guidelines. However, it is proposed a new, stepwise approach (Step 1 and 2) to treatment intensification as a tool to help physicians and patients pursue these targets in a way that fits patient profile. After Step 1, considering proceeding to the intensified goals of Step 2 is mandatory, and this intensification will be based on 10-year CVD risk, lifetime CVD risk and treatment benefit, comorbidities and patient preferences. The updated SCORE algorithm—SCORE2, SCORE-OP— is recommended in these guidelines, which estimates an individual’s 10-year risk of fatal and non-fatal CVD events (myocardial infarction, stroke) in healthy men and women aged 40–89 years. Another new and important recommendation is the use of different categories of risk according different age groups (< 50, 50−69, ≥70 years).Different flow charts of CVD risk and risk factor treatment in apparently healthy persons, in patients with established atherosclerotic CVD, and in diabetic patients are recommended. Patients with chronic kidney disease are considered high risk or very high-risk patients according to the levels of glomerular filtration rate and albumin-to-creatinine ratio. New lifestyle recommendations adapted to the ones published by the Spanish Ministry of Health as well as recommendations focused on the management of lipids, blood pressure, diabetes and chronic renal failure are included.

Multiobjective Optimization Design of Interpretable Evolutionary Fuzzy Systems With Type Self-Organizing Learning of Fuzzy Sets

This paper proposes a new multiobjective optimization approach for designing a self-generated interpretable fuzzy logic system (FLS). The types of fuzzy sets can be constructed automatically by self-organizing method, so as to form a hybrid fuzzy system. Different from the existing evolutionary type-1 fuzzy system, which is full of type-1 fuzzy sets, and the evolutionary interval type-2 fuzzy system, which is full of interval type-2 fuzzy sets, there are both type-1 fuzzy sets and interval type-2 fuzzy sets in the hybrid fuzzy system. A new transparency-oriented objective function is defined, and the constraint of the footprint of uncertainty (FOU) of the interval type-2 (IT2) fuzzy set (FS) is considered for the first time. A new FS merging criterion focusing on the proximity of the cores of fuzzy sets is proposed, which is easy to calculate and maintains the characteristics of classical similarity measures. Combined with the new merging criterion, the online cluster and fuzzy set updating (OCFU) algorithm is employed to initialize the reference rule base and the type of fuzzy sets, as it is assumed that no training data are collected in advance. Based on the reference rule base, the advanced multiobjective front-guided continuous ant colony optimization (AMO-FCACO) algorithm is introduced to optimize all the free parameters of the FLS. With the operation mentioned above, the self-generated FLSs achieve a good balance between interpretability and performance. The effectiveness of the proposed method is verified by three nonlinear system tracking problems.

Low-Rank Gradient Descent for Memory-Efficient Training of Deep In-Memory Arrays

The movement of large quantities of data during the training of a deep neural network presents immense challenges for machine learning workloads, especially those based on future functional memories deployed to store network models. As the size of network models begins to vastly outstrip traditional silicon computing resources, functional memories based on flash, resistive switches, magnetic tunnel junctions, and other technologies can store these new ultra-large models. However, new approaches are then needed to minimize hardware overhead, especially on the movement and calculation of gradient information that cannot be efficiently contained in these new memory resources. To do this, we introduce streaming batch principal component analysis (SBPCA) as an update algorithm. Streaming batch principal component analysis uses stochastic power iterations to generate a stochastic rank- k approximation of the network gradient. We demonstrate that the low-rank updates produced by streaming batch principal component analysis can effectively train convolutional neural networks on a variety of common datasets, with performance comparable to standard mini-batch gradient descent. Our approximation is made in an expanded vector form that can efficiently be applied to the rows and columns of crossbars for array-level updates. These results promise improvements in the design of application-specific integrated circuits based around large vector-matrix multiplier memories.

Parallel Network-Based Sliding Mode Tracking Control for Robotic Manipulators with Uncertain Dynamics

Robot dynamics model uncertainty and unpredictable external perturbations are important factors that influence control accuracy and stability. To accurately compensate for the dynamics model in sliding mode control (SMC), a new parallel network (PCR) is proposed in this paper. The network parallelizes the radial basis function and convolutional neural network, which gives it the advantage of making full use of one-dimensional data fitting results and two-dimensional data feature information, realizing the deep learning of multidimensional data and improving the model’s compensation accuracy and anti-interference ability. Meanwhile, based on the integration of adaptive control techniques and gradient descent, a new weight update algorithm is designed to realize the online learning of PCR networks under loss-free functions. Then, a new sliding mode controller (PCR-SMC) is established. The model-free intelligent control of the robot is accomplished without knowledge of the predetermined upper bounds. Additionally, the stability analysis of the control system is proved by the Lyapunov theorem. Lastly, robot tracking control simulations are performed on two trajectories. The results demonstrate the high-precision tracking performance of this controller in comparison with the RBF-SMC controller.

Singular value decomposition based learning identification for linear time‐varying systems: From recursion to iteration

AbstractSystem identification is a critical task in various engineering applications such as motion control, signal processing and robotics. In this article, the identification of linear time‐varying (LTV) systems that perform tasks repetitively over a finite‐time interval is investigated. Traditional LTV system identification typically adopts recursive algorithms in the time domain, which are incapable of tracking drastic‐varying parameters and are subject to estimation lag and numerical instability. To address these issues, this article proposes the utilization of an iteration axis in addition to the time axis for estimating repetitive time‐varying parameters. Specifically, the proposed approach involves an estimation algorithm for the time‐varying parameters based on a recursive least squares (RLS) method along the iteration axis, as well as an update algorithm for the covariance matrix based on singular value decomposition (SVD) to enhance numerical stability. Additionally, a bias compensation method based on noise variance estimation is introduced for the sake of eliminating estimation error induced by measurement noise. Numerical comparisons with existing methods are conducted to demonstrate the effectiveness and superiority of the proposed method.

Pediatric invasive fungal rhinosinusitis.

Pediatric invasive fungal rhinosinusitis (PIFR) is a rapidly progressive, potentially fatal disease. Previous medical literature demonstrates that its early diagnosis significantly reduces the risk of mortality in these patients. This study aims to present an updated clinical algorithm for optimized diagnosis and management of PIFR. A comprehensive review was conducted with only original, full-text articles published in English and Spanish from Cochrane Library, Pub-Med/MEDLINE, Embase, Scopus, and Google Scholar between January 2010 and June 2022. Relevant information was extracted and then integrated to develop a clinical algorithm for a proper diagnosis and management of PIFR.

Survey on Object Detection Using Deep Reinforcement Learning

Deep Reinforcement Learning (DRL) is a method that is a combination of Reinforcement Learning framework and deep neural networks. It is observed that DRL achieved a remarkable victory over the fields such as video games, robotics, finance, computer vision, health care etc. Comparing other domains, the medicine and healthcare field has benefitted a lot from DRL. In this paper, we study the role of DRL in object detection using the works of various authors. Here we focus on object detection in medicine and the healthcare field. It is observed that the authors experience higher speed in the DRL algorithm compared to classic methods. The respective methods are more efficient and accurate working on CT/MRI images. Most authors use an updated DRL algorithm in the stage of feature extraction and also club it with some machine learning techniques. DQN (Deep Q Network), Double DQN, TRPO(Trust Region Policy Optimization) etc are some common DRL algorithms used by researchers. This literature survey emphasizes methodologies of application of DRL algorithms for more efficient object detection. This review helps the futuristic way to develop a DRL algorithm for better object detection in the healthcare domain and similar ones.

Regional Algorithm for Estimating High Coccolithophore Concentration in the Northeastern Part of the Black Sea

A modified regional algorithm to quantify the coccolithophore concentration in the northeastern part of the Black Sea under conditions of intense bloom is presented. To modify the algorithm, the data of in situ measurements of coccolithophore Emiliania huxleyi abundance performed in June 2017 and 2022 (when the maximum values were 9 × 106 and 13 × 106 Cells L−1, respectively), as well as the data from hydro-optical and satellite measurements, were used. In addition, the ratio between the number of detached coccoliths and coccolithophore cells was taken into account. Based on the expanded array of in situ data, the optimal values of the regional algorithm parameters were obtained. The modified algorithm makes it possible to obtain more accurate results in areas of high coccolithophore concentrations and takes into account the contribution of coccoliths. To test the sensitivity of the algorithm to variations in bio-optical characteristics, model calculations were performed using Hydrolight software. The updated algorithm is significantly less sensitive to variations in chlorophyll concentration and CDOM absorption coefficient than its previous version.

Evaluation of the updated SOCcer v2 algorithm for coding free-text job descriptions in three epidemiologic studies.

Computer-assisted coding of job descriptions to standardized occupational classification codes facilitates evaluating occupational risk factors in epidemiologic studies by reducing the number of jobs needing expert coding. We evaluated the performance of the 2nd version of SOCcer, a computerized algorithm designed to code free-text job descriptions to US SOC-2010 system based on free-text job titles and work tasks, to evaluate its accuracy. SOCcer v2 was updated by expanding the training data to include jobs from several epidemiologic studies and revising the algorithm to account for nonlinearity and incorporate interactions. We evaluated the agreement between codes assigned by experts and the highest scoring code (a measure of confidence in the algorithm-predicted assignment) from SOCcer v1 and v2 in 14,714 jobs from three epidemiology studies. We also linked exposure estimates for 258 agents in the job-exposure matrix CANJEM to the expert and SOCcer v2-assigned codes and compared those estimates using kappa and intraclass correlation coefficients. Analyses were stratified by SOCcer score, score distance between the top two scoring codes from SOCcer, and features from CANJEM. SOCcer's v2 agreement at the 6-digit level was 50%, compared to 44% in v1, and was similar for the three studies (38%-45%). Overall agreement for v2 at the 2-, 3-, and 5-digit was 73%, 63%, and 56%, respectively. For v2, median ICCs for the probability and intensity metrics were 0.67 (IQR 0.59-0.74) and 0.56 (IQR 0.50-0.60), respectively. The agreement between the expert and SOCcer assigned codes linearly increased with SOCcer score. The agreement also improved when the top two scoring codes had larger differences in score. Overall agreement with SOCcer v2 applied to job descriptions from North American epidemiologic studies was similar to the agreement usually observed between two experts. SOCcer's score predicted agreement with experts and can be used to prioritize jobs for expert review.

Cardiovascular Health and Life Expectancy Among Adults in the United States.

Cardiovascular disease may be the main reason for stagnant growth in life expectancy in the United States since 2010. The American Heart Association recently released an updated algorithm for evaluating cardiovascular health (CVH)-Life's Essential 8 (LE8) score. We aimed to quantify the associations of CVH levels, estimated by the LE8 score, with life expectancy in a nationally representative sample of US adults. We included 23 003 nonpregnant, noninstitutionalized participants aged 20 to 79 years who participated in the National Health and Nutrition Examination Survey from 2005 to 2018 and whose mortality was identified through linkage to the National Death Index through December 31, 2019. The overall CVH was evaluated by the LE8 score (range, 0-100), as well as the score for each component of diet, physical activity, tobacco/nicotine exposure, sleep duration, body mass index, non-high-density lipoprotein cholesterol, blood glucose, and blood pressure. Life table method was used to estimate life expectancy by levels of the CVH. During a median of 7.8 years of follow-up, 1359 total deaths occurred. The estimated life expectancy at age 50 years was 27.3 years (95% CI, 26.1-28.4), 32.9 years (95% CI, 32.3-33.4), and 36.2 years (95% CI, 34.2-38.2) in participants with low (LE8 score <50), moderate (50≤ LE8 score <80), and high (LE8 score ≥80) CVH, respectively. Equivalently, participants with high CVH had an average 8.9 (95% CI, 6.2-11.5) more years of life expectancy at age 50 years compared with those with low CVH. On average, 42.6% of the gained life expectancy at age 50 years from adhering to high CVH was attributable to reduced cardiovascular disease death. Similarly significant associations of CVH with life expectancy were observed in men and women, respectively. Similarly significant associations of CVH with life expectancy were observed in White participants and Black participants but not in Mexican participants. Adhering to a high CVH, defined as the LE8 score, is related to a considerably increased life expectancy in US adults, but more research needs to be done in other races and ethnicities (eg, Hispanic and Asian).

B-cell lineage neoplasms transdifferentiating into histiocytic/dendritic cell neoplasms: diversity, differentiation lineage, genomic alterations, and therapy: Report from the 2021 SH/EAHP Workshop

To report findings from the 2021 Society for Hematopathology/European Association for Haematopathology Workshop within the category of B-cell lineage neoplasms' transdifferentiation into histiocytic/dendritic cell neoplasms (HDCNs). The workshop panel reviewed 29 cases, assigned consensus diagnoses, and summarized findings. The specific diagnoses of transdifferentiated HDCN tumors were histiocytic sarcoma (16); Langerhans cell histiocytosis/sarcoma (5); indeterminate DC tumor (1); and HDCN, unclassifiable (1). Approximately one-third of the patients reviewed had follicular lymphoma; lymphoblastic leukemia/lymphoma; or another B-cell lymphoma, most commonly chronic lymphocytic leukemia/small lymphocytic lymphoma. There was a 3:1 preponderance toward women, median patient age was 60 years, and the median interval between the initial diagnosis of B-cell lineage neoplasm and HDCN was 4 to 5 years. The submitted cases have demonstrated substantial heterogeneity as well as overlapping immunophenotypic and other features. Comprehensive genomic DNA sequencing revealed alterations enriched in the MAPK pathway. Based on shared and distinct alterations seen in HDCNs and the preceding lymphomas, both linear and divergent clonal evolutionary pathways were inferred. Furthermore, RNA sequencing performed in a subset of cases yielded new insights into markers that could be useful for more precise cell lineage identification. The panel has thus proposed an updated algorithm for HDCN lineage assignment. The outcome of transdifferentiated HDCNs was poor, but the MAPK signaling pathway emerges as a potentially attractive therapeutic target. Transdifferentiated HDCNs demonstrate heterogeneity and pose diagnostic challenges with regard to exact classification, but the in-depth characterization of the submitted cases have added to our understanding of the secondary HDCNs transdifferentiated from B-cell lymphoma/leukemia. Continuous efforts focusing on deciphering the specific cell lineage and differentiation state of these tumors will be critical for their accurate classification. Comprehensive molecular characterization of HDCNs may be informative in this regard. With the list of novel pharmacologic inhibitors of the MAPK pathway continuing to expand, improved outcomes for HDCN can be expected.

Delayless partial subband update algorithm for feed-forward active road noise control system in pure electric vehicles

Road noise is one of the dominant noise sources in pure electric vehicles. Active control for road noise is an attractive solution since this technique is suitable for controlling low-frequency noise. Feed-forward active road noise control (ARNC) systems utilize lots of reference signals and corresponding adaptive filters, and the control performance usually improves as the number of reference signals increases. To reduce the computational cost of ARNC systems so that more reference signals can be processed, we introduce the delayless subband structure into ARNC system and a delayless partial update subband with weighted reference signal normalized filtered-x least mean square (DPUS-WRS-NFxLMS) algorithm is proposed in this paper. Based on the traditional delayless subband algorithm, a partial subband update strategy is proposed to reduce the computational cost. Furthermore, the weighted reference signal is introduced for each subband filter. In this way, a modified vector is designed to assign different weights to different reference signal samples so that the convergence performance can be improved and the round-off error can be reduced. In addition, an adjustment strategy for the step size and reference signal weights is designed to make the algorithm have a better balance between the convergence rate and steady-state behavior. Finally, a purely time-delayed secondary path is constructed as the estimated secondary path to avoid the negative impact of the magnitude-frequency characteristics of the actual secondary path. The simulation results prove that the proposed algorithm achieves better control results than the traditional algorithms with lower computational complexity.

Traffic-aware efficient consistency update in NFV-enabled software defined networking

In network function virtualization(NFV)-enabled software defined networks, the controller needs to frequently update the flow forwarding rules in the data plane to adapt to dynamic changes in network topologies or service requests. However, inconsistent rule updates may lead to blackholes, loops, transient congestion or policy violations (e.g., packets do not traverse designated network functions in a specific order), resulting in service interruption and throughput degradation. Therefore, this paper proposes an effective rule consistent update mechanism to avoid the above four problems simultaneously, while improving network throughput and satisfying user requests. Specifically, we first build three effective models to avoid blackholes, loops, and policy violations. Then, considering that network function nodes may change the sizes of their processed flows, we build a congestion avoidance model based on traffic changes to avoid congestion, which can reduce unnecessary rule update delays and packet loss. Subsequently, we prove that the consistent update problem constructed above is NP-hard, and then design an effective heuristic rule consistent update algorithm to obtain the rule update sequence that can simultaneously avoid blackholes, loops, congestion, and policy violations. Extensive trace-driven simulation results show that compared with the existing update methods, our proposed method can improve the success rate by up to 20.6% and reduce the maximum link utilization by up to 7.5%.

Abstract 4398: Validation of an updated algorithm to identify non-small cell lung cancer (NSCLC) patients in administrative claims databases

Abstract Purpose: Administrative claims data using the ICD-9-CM or ICD-10-CM coding systems are not detailed enough to distinguish between lung cancer subtypes, which presents challenges for real-world research. This study updated and validated a previous treatment-based algorithm that uses claims data to identify NSCLC vs small-cell lung cancer (SCLC) (Turner RM, et al. Front Pharmacol 2017;8:883). Methods: This study used Optum’s de-identified Market Clarity Data (2007-2021) which deterministically links medical and pharmacy claims with electronic health record (EHR) data from providers across the continuum of care. Included patients had lung cancer diagnosis and histology information available in the EHR, which is the gold standard for validation. Patients were required to have ≥3 months of continuous health plan enrollment before and after the diagnosis date. First, to replicate the Turner algorithm, patients were identified between Jun 2014 and Oct 2015 with any first-line NSCLC treatment and no SCLC treatment (Step 1). Next, the Turner algorithm was applied from Nov 2015 to Dec 2020 to evaluate if performance decreased given the approval of immunotherapies since 2015 (Step 2). Finally, the Turner algorithm was updated with NSCLC treatments approved since 2015 in concordance with the latest US treatment guidelines. Algorithm performance at each step was measured using sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Results: Sample sizes and performance statistics for each step are presented in the Table. From Step 1 to Step 2, specificity and PPV decreased while sensitivity and NPV increased. Algorithm performance improved slightly for all measures from Step 2 to Step 3. Conclusions: This updated treatment-based algorithm improves validity and accuracy in identifying NSCLC patients in claims databases and supports its use in real-world research using claims databases when histology data is unavailable. Step 1: Turner replication Step 2: Turner 2015-2020 Step 3: Updated Turner 2015-2020 Date range Jun 2014-Oct 2015 Nov 2015-Dec 2020 Nov 2015-Dec 2020 Sample size 406 2573 2744 Sensitivity 0.873 0.920 0.932 Specificity 0.933 0.865 0.923 PPV 0.987 0.976 0.988 NPV 0.560 0.640 0.673 Citation Format: Sandip Pravin Patel, Rongrong Wang, Summera Qiheng Zhou, Daniel Sheinson, Ann Johnson, Janet Lee. Validation of an updated algorithm to identify non-small cell lung cancer (NSCLC) patients in administrative claims databases. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4398.

Scalable NMF via linearly optimized data compression.

Orthonormal projective non-negative matrix factorization (opNMF) has been widely used in neuroimaging and clinical neuroscience applications to derive representations of the brain in health and disease. The non-negativity and orthonormality constraints of opNMF result in intuitive and well-localized factors. However, the advantages of opNMF come at a steep computational cost that prohibits its use in large-scale data. In this work, we propose novel and scalable optimization schemes for orthonormal projective non-negative matrix factorization that enable the use of the method in large-scale neuroimaging settings. We replace the high-dimensional data matrix with its corresponding singular value decomposition (SVD) and QR decompositions and combine the decompositions with opNMF multiplicative update algorithm. Empirical validation of the proposed methods demonstrated significant speed-up in computation time while keeping memory consumption low without compromising the accuracy of the solution.

The role of digital tomosynthesis in laryngeal cancer: comparison with radiography and computed tomography

Digital tomosynthesis is a relatively new imaging modality that is already used in the diagnosis of breast cancer and has shown promising results in evaluating patients with pulmonary, osteoarticular, and other pathologies. However, up to date, there are no published studies related to the usefulness of digital tomosynthesis in the evaluation of patients with laryngeal cancer. The purpose of this study was to evaluate the clinical role of digital tomosynthesis in the diagnosis of laryngeal cancer and compare the imaging results with those obtained by digital radiography and computed tomography. Material and methods. The study was carried out between 2015 and 2019 at the Institute of Oncology in the Republic of Moldova and included 253 consecutive patients with laryngeal cancer referred to the Institute of Oncology during this period. All patients underwent digital radiography and digital tomosynthesis investigations. In 41 patients who provided written informed consent, computed tomography was performed. The results of all imaging investigations were compared. Results. The statistical analysis revealed a high degree of agreement and a strong linear correlation between the data obtained with digital tomosynthesis and computed tomography, as well as concordance correlation coefficients for different parameters between 0.63 and 1.0 (mean value = 0.82±0.11). For comparison, the concordance correlation coefficients for the same parameters obtained for digital radiography versus computed tomography ranged between 0.08 and 0.93 (mean value = 0.43±0.25). An updated imaging algorithm that includes digital tomosynthesis has also been proposed for investigating patients with suspected laryngeal cancer. Conclusions. The study demonstrated the usefulness of digital tomosynthesis for the evaluation of patients with laryngeal cancer. When compared to computed tomography, which is considered the gold standard, digital tomosynthesis revealed a much higher performance compared to digital radiography. Considering the availability of low-dose protocols for digital tomosynthesis, the modality might also be helpful for laryngeal cancer screening in a high-risk population. However, new studies are also required to confirm our findings and define the place of digital tomosynthesis in the imaging algorithm for patients with laryngeal cancer.

Fast and Accurate Non-Negative Latent Factor Analysis of High-Dimensional and Sparse Matrices in Recommender Systems

A fast non-negative latent factor (FNLF) model for a high-dimensional and sparse (HiDS) matrix adopts a Single Latent Factor-dependent, Non-negative, Multiplicative and Momentum-incorporated Update (SLF-NM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> U) algorithm, which enables its fast convergence. It is crucial to achieve a rigorously theoretical proof regarding its fast convergence, which has not been provided in prior research. Aiming at addressing this critical issue, this work theoretically proves that with an appropriately chosen momentum coefficient, SLF-NM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> U enables the fast convergence of an FNLF model in both continuous and discrete time cases. Empirical analysis of HiDS matrices generated by representative industrial applications provides empirical evidences for the theoretical proof. Hence, this study represents an important milestone in the field of HiDS matrix analysis.