Customized deep learning based Turkish automatic speech recognition system supported by language model.

Background: In today's world, numerous applications integral to various facets of daily life include automatic speech recognition methods. Thus, the development of a successful automatic speech recognition system can significantly augment the convenience of people's daily routines. While many automatic speech recognition systems have been established for widely spoken languages like English, there has been insufficient progress in developing such systems for less common languages such as Turkish. Moreover, due to its agglutinative structure, designing a speech recognition system for Turkish presents greater challenges compared to other language groups. Therefore, our study focused on proposing deep learning models for automatic speech recognition in Turkish, complemented by the integration of a language model. Methods: In our study, deep learning models were formulated by incorporating convolutional neural networks, gated recurrent units, long short-term memories, and transformer layers. The Zemberek library was employed to craft the language model to improve system performance. Furthermore, the Bayesian optimization method was applied to fine-tune the hyper-parameters of the deep learning models. To evaluate the model's performance, standard metrics widely used in automatic speech recognition systems, specifically word error rate and character error rate scores, were employed. Results: Upon reviewing the experimental results, it becomes evident that when optimal hyper-parameters are applied to models developed with various layers, the scores are as follows: Without the use of a language model, the Turkish Microphone Speech Corpus dataset yields scores of 22.2 -word error rate and 14.05-character error rate, while the Turkish Speech Corpus dataset results in scores of 11.5 -word error rate and 4.15 character error rate. Upon incorporating the language model, notable improvements were observed. Specifically, for the Turkish Microphone Speech Corpus dataset, the word error rate score decreased to 9.85, and the character error rate score lowered to 5.35. Similarly, the word error rate score improved to 8.4, and the character error rate score decreased to 2.7 for the Turkish Speech Corpus dataset. These results demonstrate that our model outperforms the studies found in the existing literature.

Age-Invariant Genes: Multi-Tissue Identification and Characterization of Murine Reference Genes.

Studies of the aging transcriptome focus on genes that change with age. But what can we learn from age-invariant genes-those that remain unchanged throughout the aging process? These genes also have a practical application: they serve as reference genes (often called housekeeping genes) in expression studies. Reference genes have mostly been identified and validated in young organisms, and no systematic investigation has been done across the lifespan. Here, we build upon a common pipeline for identifying reference genes in RNA-seq datasets to identify age-invariant genes across seventeen C57BL/6 mouse tissues (brain, lung, bone marrow, muscle, white blood cells, heart, small intestine, kidney, liver, pancreas, skin, brown, gonadal, marrow, and subcutaneous adipose tissue) spanning 1 to 21+ months of age. We identify 9 pan-tissue age-invariant genes and many tissue-specific age-invariant genes. These genes are stable across the lifespan and are validated in independent bulk RNA-seq datasets and RT-qPCR. We find age-invariant genes have shorter transcripts on average and are enriched for CpG islands. Interestingly, pathway enrichment analysis for age-invariant genes identifies an overrepresentation of molecular functions associated with some, but not all, hallmarks of aging. Thus, though hallmarks of aging typically involve changes in cell maintenance mechanisms, select genes associated with these hallmarks resist fluctuations in expression with age. Finally, our analysis concludes no classical reference gene is appropriate for aging studies in all tissues. Instead, we provide tissue-specific and pan-tissue genes for assays utilizing reference gene normalization (i.e., RT-qPCR) that can be applied to animals across the lifespan.

Identification of hypoxic macrophages in glioblastoma with therapeutic potential for vasculature normalization.

Monocyte-derived tumor-associated macrophages (Mo-TAMs) intensively infiltrate diffuse gliomas with remarkable heterogeneity. Using single-cell transcriptomics, we chart a spatially resolved transcriptional landscape of Mo-TAMs across 51 patients with isocitrate dehydrogenase (IDH)-wild-type glioblastomas or IDH-mutant gliomas. We characterize a Mo-TAM subset that is localized to the peri-necrotic niche and skewed by hypoxic niche cues to acquire a hypoxia response signature. Hypoxia-TAM destabilizes endothelial adherens junctions by activating adrenomedullin paracrine signaling, thereby stimulating a hyperpermeable neovasculature that hampers drug delivery in glioblastoma xenografts. Accordingly, genetic ablation or pharmacological blockade of adrenomedullin produced by Hypoxia-TAM restores vascular integrity, improves intratumoral concentration of the anti-tumor agent dabrafenib, and achieves combinatorial therapeutic benefits. Increased proportion of Hypoxia-TAM or adrenomedullin expression is predictive of tumor vessel hyperpermeability and a worse prognosis of glioblastoma. Our findings highlight Mo-TAM diversity and spatial niche-steered Mo-TAM reprogramming in diffuse gliomas and indicate potential therapeutics targeting Hypoxia-TAM to normalize tumor vasculature.

Identifying key mental health and improvement factors in hospital administrators working from home using a DEMATEL-based network analysis model.

Purpose: To identify the key mental health and improvement factors in hospital administrators working from home during COVID-19 normalization prevention and control. Methods: The survey was conducted from May to June 2023, and the practical experiences of 33 hospital administrators were collected using purposive sampling. The study examined a set of mental health factor systems. The relationship structure between the factors was constructed using the Decision-making Trial and Evaluation Laboratory (DEMATEL) method. Finally, the structure was transformed using the influence weight of each factor via the DEMATEL-based Analytic Network Process. Results: Regarding influence weight, the key mental health factors of hospital administrators are mainly "lack of coordination," "time management issues," and "work-life imbalances." The influential network relation map shows that improvements can be made by addressing "improper guidelines," "laziness due to being at home," and "job insecurity" because they are the main sources of influence. The reliability level of the results for the network structure and weight was 98.79% (i.e., the gap was 1.12% < 5%). Conclusion: The network analysis model based on DEMATEL proposed in this study can evaluate the mental health factors of hospital administrators during the pandemic period from a multidimensional and multidirectional perspective and may help improve mental health problems and provide suggestions for hospital administrators.

Deep learning-based identification of esophageal cancer subtypes through analysis of high-resolution histopathology images.

Esophageal cancer (EC) remains a significant health challenge globally, with increasing incidence and high mortality rates. Despite advances in treatment, there remains a need for improved diagnostic methods and understanding of disease progression. This study addresses the significant challenges in the automatic classification of EC, particularly in distinguishing its primary subtypes: adenocarcinoma and squamous cell carcinoma, using histopathology images. Traditional histopathological diagnosis, while being the gold standard, is subject to subjectivity and human error and imposes a substantial burden on pathologists. This study proposes a binary class classification system for detecting EC subtypes in response to these challenges. The system leverages deep learning techniques and tissue-level labels for enhanced accuracy. We utilized 59 high-resolution histopathological images from The Cancer Genome Atlas (TCGA) Esophageal Carcinoma dataset (TCGA-ESCA). These images were preprocessed, segmented into patches, and analyzed using a pre-trained ResNet101 model for feature extraction. For classification, we employed five machine learning classifiers: Support Vector Classifier (SVC), Logistic Regression (LR), Decision Tree (DT), AdaBoost (AD), Random Forest (RF), and a Feed-Forward Neural Network (FFNN). The classifiers were evaluated based on their prediction accuracy on the test dataset, yielding results of 0.88 (SVC and LR), 0.64 (DT and AD), 0.82 (RF), and 0.94 (FFNN). Notably, the FFNN classifier achieved the highest Area Under the Curve (AUC) score of 0.92, indicating its superior performance, followed closely by SVC and LR, with a score of 0.87. This suggested approach holds promising potential as a decision-support tool for pathologists, particularly in regions with limited resources and expertise. The timely and precise detection of EC subtypes through this system can substantially enhance the likelihood of successful treatment, ultimately leading to reduced mortality rates in patients with this aggressive cancer.

The role of vasculature and angiogenesis in respiratory diseases.

In European countries, nearly 10% of all hospital admissions are related to respiratory diseases, mainly chronic life-threatening diseases such as COPD, pulmonary hypertension, IPF or lung cancer. The contribution of blood vessels and angiogenesis to lung regeneration, remodeling and disease progression has been increasingly appreciated. The vascular supply of the lung shows the peculiarity of dual perfusion of the pulmonary circulation (vasa publica), which maintains a functional blood-gas barrier, and the bronchial circulation (vasa privata), which reveals a profiled capacity for angiogenesis (namely intussusceptive and sprouting angiogenesis) and alveolar-vascular remodeling by the recruitment of endothelial precursor cells. The aim of this review is to outline the importance of vascular remodeling and angiogenesis in a variety of non-neoplastic and neoplastic acute and chronic respiratory diseases such as lung infection, COPD, lung fibrosis, pulmonary hypertension and lung cancer.

Navigating tumor angiogenesis: therapeutic perspectives and myeloid cell regulation mechanism.

Sustained angiogenesis stands as a hallmark of cancer. The intricate vascular tumor microenvironment fuels cancer progression and metastasis, fosters therapy resistance, and facilitates immune evasion. Therapeutic strategies targeting tumor vasculature have emerged as transformative for cancer treatment, encompassing anti-angiogenesis, vessel normalization, and endothelial reprogramming. Growing evidence suggests the dynamic regulation of tumor angiogenesis by infiltrating myeloid cells, such as macrophages, myeloid-derived suppressor cells (MDSCs), and neutrophils. Understanding these regulatory mechanisms is pivotal in paving the way for successful vasculature-targeted cancer treatments. Therapeutic interventions aimed to disrupt myeloid cell-mediated tumor angiogenesis may reshape tumor microenvironment and overcome tumor resistance to radio/chemotherapy and immunotherapy.

Bispecific fibrous glue synergistically boosts vascular normalization and antitumor immunity for advanced renal carcinoma therapy.

Immune checkpoint blockade therapy represented by programmed cell death ligand 1 (PD-L1) inhibitor for advanced renal carcinoma with an objective response rate (ORR) in patients is less than 20%. It is attributed to abundant tumoral vasculature with abnormal structure limiting effector T cell infiltration and drug penetration. We propose a bispecific fibrous glue (BFG) to regulate tumor immune and vascular microenvironments simultaneously. The bispecific precursor glue peptide-1 (pre-GP1) can penetrate tumor tissue deeply and self-assemble into BFG in the presence of neuropilin-1 (NRP-1) and PD-L1. The resultant fibrous glue is capable of normalizing tumoral vasculature as well as restricting immune escape. The pre-GP1 retains a 6-fold higher penetration depth than that of antibody in the multicellular spheroids (MCSs) model. It also shows remarkable tumor growth inhibition (TGI) from 19% to 61% in a murine advanced large tumor model compared to the clinical combination therapy. In addition, in the orthotopic renal tumor preclinical model, the lung metastatic nodules are reduced by 64% compared to the clinically used combination. This pre-GP1 provides a promising strategy to control the progression and metastasis of advanced renal carcinoma.

Analysis of Serum Free Thyroxine Concentrations in Healthy Term Neonates Underlines Need for Local and Laboratory-Specific Reference Interval: A Systematic Review and Meta-Analysis of Individual Participant Data.

Background: Initial evaluation of the hypothalamus-pituitary-thyroid axis is done by measuring serum free thyroxine (fT4) and thyrotropin concentrations. For correct interpretation of these measurements, reliable age-specific reference intervals (RIs) are fundamental. Since neonatal fT4 RIs conforming to the Clinical and Laboratory Standards Institute guidelines are not available for all assays, we set out to create literature-based uniform age-specific neonatal fT4 RIs that may be used for every assay. Methods: For meta-analysis of individual participant fT4 concentrations, we systematically searched MEDLINE and Embase (search date December 6, 2023; PROSPERO registration CRD42016041871). We searched for studies reporting fT4 concentrations in healthy term newborns aged 2-27 days, born to mothers without thyroid disease in iodine-sufficient regions. Authors were invited to supply data. Due to standardization differences between assays, data could not be combined for meta-analysis directly, and we attempted to normalize the data using two distinct methods. Results: We obtained 4206 fT4 concentrations from 20 studies that used 13 different assays from 6 manufacturers. First, we set out to normalize fT4 data using the mean and standard deviation of (assay-specific) adult RIs. fT4 concentrations were transformed into Z-scores, assuming a normal distribution. Using a linear mixed-effects model (LMM), we still found a significant difference between fT4 concentration across studies (p < 0.001), after this normalization. As a second approach, we normalized the fT4 concentrations using data from a method/assay comparison study. We used the relationship between the Cobas assay and the other assays as a reference point to convert all values to Cobas values. However, this method also failed to produce consistent results, with significant differences between the normalized data (LMM p < 0.001). Conclusions: We conclude that our attempts at normalizing fT4 assay results were unsuccessful. Confounders related to our unsuccessful analysis may be assay related and/or biological. These findings have significant implications for patient care, since relying on RIs from literature may result in erroneous interpretation of results. Therefore, we strongly recommend to establish local RIs for accurate interpretation of serum fT4 concentrations in neonates.

Chitosan-based hybrid nanospheres for vessel normalization towards enhancing tumor chemotherapy.

Vessel normalization has proved imperative in tumor growth inhibition. In this work, biopolymer-based hybrid nanospheres capable of normalizing blood vessels were designed to improve the therapeutic effect of chemotherapeutic drugs. Zn0.4Fe2.6O4 nanoparticles (ZFO NPs) were synthesized, and were encapsulated in cross-inked chitosan (CS) along with a nitric oxide (NO) precursor, DETA NONOate, forming hybrid ZFO/NO@CS nanospheres highly stable in physiological environment. The structure, morphology and size of the nanospheres were characterized. The ZFO/NO@CS nanospheres could release NO under acidic conditions typical of intratumoral and intracellular environment. The results of related factors expression, wound healing and tube formation assays demonstrated that both the encapsulated ZFO NPs and the released NO were able to inhibit angiogenesis in tumors. The ZFO/NO@CS nanospheres enhanced the antitumor efficacy of the chemotherapeutic drug DOX by normalizing tumor vessels, as evidenced by in vivo experiments for CT26 tumor-bearing mice. By analyzing the contents of Fe in the tumor and different organs, the nanospheres were found to accumulate primarily at the tumor site. The blood analysis showed little side effect of the nanospheres. The ZFO/NO@CS nanospheres have great potential in improving tumor therapeutic effect when used in combination with chemotherapeutic drugs.

Perceptual Resolution of Ambiguity: Can Tuned, Divisive Normalization Account for both Interocular Similarity Grouping and Difference Enhancement.

Our visual system usually provides a unique and functional representation of the external world. At times, however, the visual system has more than one compelling interpretation of the same retinal stimulus; in this case, neural populations compete for perceptual dominance to resolve ambiguity. Spatial and temporal context can guide perceptual experience. Recent evidence shows that ambiguous retinal stimuli are sometimes resolved by enhancing either similarity or differences among multiple percepts. Divisive normalization is a canonical neural computation that enables context-dependent sensory processing by attenuating a neuron's response by other neurons. Experiments here show that divisive normalization can account for perceptual representations of either similarity enhancement (so-called grouping) or difference enhancement, offering a unified framework for opposite perceptual outcomes.

lute: estimating the cell composition of heterogeneous tissue with varying cell sizes using gene expression.

Relative cell type fraction estimates in bulk RNA-sequencing data are important to control for cell composition differences across heterogenous tissue samples. Current computational tools estimate relative RNA abundances rather than cell type proportions in tissues with varying cell sizes, leading to biased estimates. We present lute, a computational tool to accurately deconvolute cell types with varying sizes. Our software wraps existing deconvolution algorithms in a standardized framework. Using simulated and real datasets, we demonstrate how lute adjusts for differences in cell sizes to improve the accuracy of cell composition. Software is available from https://bioconductor.org/packages/lute.

A bibliometric analysis of the research landscape on vascular normalization in cancer.

Tumor vascular normalization profoundly affects the advancement of cancer therapy. Currently, with the rapid increase in research on tumor vascular normalization, few analytical and descriptive studies have investigated the trends in its development, key research power, present research hotspots, and future outlooks. In this study, articles and reviews published between January 1, 2003, and October 29, 2022 were retrieved from Web of Science database. Subsequently, published research trends, countries/regions, institutions, authors, journals, references, and keywords were analyzed based on traditional bibliometric laws (such as Price's exponential growth, Bradford's, Lotka's, and Zipf's). Our results showed that the last two decades have seen an increase in tumor vascular normalization research. USA emerged as the preeminent contributor to the field, boasting the highest H-index and accruing the greatest quantity of publications and citations. Among institutions, Massachusetts General Hospital and Harvard University made significant contributions, and Professor RK Jain was identified as a key leader in this field. Out of 583 academic journals, Cancer Research and Clinical Cancer Research published the most articles on vascular normalization. The research focal points in the field primarily include immunotherapy, tumor microenvironments, nanomedicine, and emerging frontier themes such as metabolism and mechanomedicine. Concurrently, the challenges of vascular normalization in cancer are discussed as well. In conclusion, the study presented a thorough analysis of the literature covering the past 20 years on vascular normalization in cancer, highlighting leading countries, institutions, authors, journals, and the emerging research focal points in this field. Future studies will advance the ongoing efforts in the field of tumor vascular normalization, aiming to enhance our ability to effectively manage and treat cancer.

Physical activity on prescription for children with obesity: a focus group study exploring experiences in paediatric healthcare.

Background: Insufficient physical activity is a growing public health concern and is closely linked to obesity in both adults and children. Swedish physical activity on prescription (PAP) is effective in increasing physical activity levels in adults, but knowledge about how PAP is used in paediatric healthcare is lacking. Therefore, this study aimed to explore experiences of working with PAP for children with obesity amongst paediatric staff and managers. Methods: Seven focus group discussions with 26 participants from paediatric outpatient clinics in western Sweden were conducted. Data were analysed both inductively and deductively, framed by the Normalization Process Theory's four core constructs: coherence, cognitive participation, collective action, and reflexive monitoring. Results: The PAP work for children with obesity was experienced to be about helping children to become physically active, and less about losing weight. Identified barriers for using PAP were the non-uniform nature of the work and a perceived lack of guidelines. Collaboration with physiotherapists and physical activity organisers outside the organisation was identified as an important facilitator. An important contextual factor for implementing PAP is the collaboration between paediatric clinics and physical activity organisers. In the transition between these stakeholders, maintaining a family-centred approach when working with PAP was experienced as challenging. Conclusions: PAP is a well-known intervention that is inconsistently used for children with obesity. The intervention should include a family-centred approach for this patient group. It also needs to align better with existing collaborations with other healthcare units as well as with new forms of collaboration with physical activity organisers in the community.

MKG-GC: A multi-task learning-based knowledge graph construction framework with personalized application to gastric cancer.

Over the past decade, information for precision disease medicine has accumulated in the form of textual data. To effectively utilize this expanding medical text, we proposed a multi-task learning-based framework based on hard parameter sharing for knowledge graph construction (MKG), and then used it to automatically extract gastric cancer (GC)-related biomedical knowledge from the literature and identify GC drug candidates. In MKG, we designed three separate modules, MT-BGIPN, MT-SGTF and MT-ScBERT, for entity recognition, entity normalization, and relation classification, respectively. To address the challenges posed by the long and irregular naming of medical entities, the MT-BGIPN utilized bidirectional gated recurrent unit and interactive pointer network techniques, significantly improving entity recognition accuracy to an average F1 value of 84.5% across datasets. In MT-SGTF, we employed the term frequency-inverse document frequency and the gated attention unit. These combine both semantic and characteristic features of entities, resulting in an average Hits@ 1 score of 94.5% across five datasets. The MT-ScBERT integrated cross-text, entity, and context features, yielding an average F1 value of 86.9% across 11 relation classification datasets. Based on the MKG, we then developed a specific knowledge graph for GC (MKG-GC), which encompasses a total of 9129 entities and 88,482 triplets. Lastly, the MKG-GC was used to predict potential GC drugs using a pre-trained language model called BioKGE-BERT and a drug-disease discriminant model based on CNN-BiLSTM. Remarkably, nine out of the top ten predicted drugs have been previously reported as effective for gastric cancer treatment. Finally, an online platform was created for exploration and visualization of MKG-GC at https://www.yanglab-mi.org.cn/MKG-GC/.

Impact of Translation on Biomedical Information Extraction: Experiment on Real-Life Clinical Notes.

Background: Biomedical natural language processing tasks are best performed with English models, and translation tools have undergone major improvements. On the other hand, building annotated biomedical data sets remains a challenge. Objective: The aim of our study is to determine whether the use of English tools to extract and normalize French medical concepts based on translations provides comparable performance to that of French models trained on a set of annotated French clinical notes. Methods: We compared 2 methods: 1 involving French-language models and 1 involving English-language models. For the native French method, the named entity recognition and normalization steps were performed separately. For the translated English method, after the first translation step, we compared a 2-step method and a terminology-oriented method that performs extraction and normalization at the same time. We used French, English, and bilingual annotated data sets to evaluate all stages (named entity recognition, normalization, and translation) of our algorithms. Results: The native French method outperformed the translated English method, with an overall F1-score of 0.51 (95% CI 0.47-0.55), compared with 0.39 (95% CI 0.34-0.44) and 0.38 (95% CI 0.36-0.40) for the 2 English methods tested. Conclusions: Despite recent improvements in translation models, there is a significant difference in performance between the 2 approaches in favor of the native French method, which is more effective on French medical texts, even with few annotated documents.

Multi-domain stain normalization for digital pathology: A cycle-consistent adversarial network for whole slide images.

The variation in histologic staining between different medical centers is one of the most profound challenges in the field of computer-aided diagnosis. The appearance disparity of pathological whole slide images causes algorithms to become less reliable, which in turn impedes the wide-spread applicability of downstream tasks like cancer diagnosis. Furthermore, different stainings lead to biases in the training which in case of domain shifts negatively affect the test performance. Therefore, in this paper we propose MultiStain-CycleGAN, a multi-domain approach to stain normalization based on CycleGAN. Our modifications to CycleGAN allow us to normalize images of different origins without retraining or using different models. We perform an extensive evaluation of our method using various metrics and compare it to commonly used methods that are multi-domain capable. First, we evaluate how well our method fools a domain classifier that tries to assign a medical center to an image. Then, we test our normalization on the tumor classification performance of a downstream classifier. Furthermore, we evaluate the image quality of the normalized images using the Structural similarity index and the ability to reduce the domain shift using the Fréchet inception distance. We show that our method proves to be multi-domain capable, provides a very high image quality among the compared methods, and can most reliably fool the domain classifier while keeping the tumor classifier performance high. By reducing the domain influence, biases in the data can be removed on the one hand and the origin of the whole slide image can be disguised on the other, thus enhancing patient data privacy.

Quantification Quality Control Emerges as A Crucial Factor to Enhance Single Cell Proteomics Data Analysis.

Mass spectrometry (MS)-based single-cell proteomics (SCP) provides us the opportunity to unbiasedly explore biological variability within cells without the limitation of antibody availability. This field is rapidly developed with the main focuses on instrument advancement, sample preparation refinement, and signal boosting methods; however, the optimal data processing and analysis are rarely investigated which holds an arduous challenge because of the high proportion of missing values and batch effect. Here, we introduced a quantification quality control to intensify the identification of differentially expressed proteins (DEPs) by considering both within and across SCP data. Combining quantification quality control with isobaric matching between runs (IMBR) and PSM-level normalization, an additional 12% and 19% of proteins and peptides, with more than 90% of proteins/peptides containing valid values, were quantified. Clearly, quantification quality control was able to reduce quantification variations and q-values with the more apparent cell type separations. In addition, we found that PSM-level normalization performed similarly to other protein-level normalizations but kept the original data profiles without the additional requirement of data manipulation. In proof of concept of our refined pipeline, six uniquely identified DEPs exhibiting varied fold-changes and playing critical roles for melanoma and monocyte functionalities were selected for validation using immunoblotting. Five out of six validated DEPs showed an identical trend with the SCP dataset, emphasizing the feasibility of combining the IMBR, cell quality control, and PSM-level normalization in SCP analysis, which is beneficial for future SCP studies.

LIRRN: Location-Independent Relative Radiometric Normalization of Bitemporal Remote-Sensing Images.

Relative radiometric normalization (RRN) is a critical pre-processing step that enables accurate comparisons of multitemporal remote-sensing (RS) images through unsupervised change detection. Although existing RRN methods generally have promising results in most cases, their effectiveness depends on specific conditions, especially in scenarios with land cover/land use (LULC) in image pairs in different locations. These methods often overlook these complexities, potentially introducing biases to RRN results, mainly because of the use of spatially aligned pseudo-invariant features (PIFs) for modeling. To address this, we introduce a location-independent RRN (LIRRN) method in this study that can automatically identify non-spatially matched PIFs based on brightness characteristics. Additionally, as a fast and coregistration-free model, LIRRN complements keypoint-based RRN for more accurate results in applications where coregistration is crucial. The LIRRN process starts with segmenting reference and subject images into dark, gray, and bright zones using the multi-Otsu threshold technique. PIFs are then efficiently extracted from each zone using nearest-distance-based image content matching without any spatial constraints. These PIFs construct a linear model during subject-image calibration on a band-by-band basis. The performance evaluation involved tests on five registered/unregistered bitemporal satellite images, comparing results from three conventional methods: histogram matching (HM), blockwise KAZE, and keypoint-based RRN algorithms. Experimental results consistently demonstrated LIRRN's superior performance, particularly in handling unregistered datasets. LIRRN also exhibited faster execution times than blockwise KAZE and keypoint-based approaches while yielding results comparable to those of HM in estimating normalization coefficients. Combining LIRRN and keypoint-based RRN models resulted in even more accurate and reliable results, albeit with a slight lengthening of the computational time. To investigate and further develop LIRRN, its code, and some sample datasets are available at link in Data Availability Statement.