Background: The assessment of sperm DNA integrity has been proposed as a complementary test to conventional mammalian semen analysis. In this sense, single-strand (SSB) and double-strand (DSB) DNA breaks, the two types of sperm DNA fragmentation (SDF), have been reported to have different aetiologies and to be associated to different fertility outcomes in bovine and humans. Considering that no studies in porcine have addressed how SDF may affect sperm quality and fertility outcomes, the present work aimed to determine the impact of global DNA damage, SSB and DSB on sperm quality and in vitro fertilising ability. To this end, 24 ejaculates (one per boar) were split into three aliquots: the first was used to assess sperm quality parameters through a computer-assisted sperm analysis (CASA) system and flow cytometry; the second was used to perform in vitro fertilisation, and the third, to evaluate sperm DNA integrity using alkaline and neutral Comet assays. Results: The results showed that global DNA damage negatively correlates (P < 0.05) with normal sperm morphology (R = - 0.460) and progressive motility (R = - 0.419), and positively with the percentage of non-viable sperm (R = 0.507). Multiple regression analyses showed that non-viable sperm were related to SSB (β = - 0.754). In addition, while fertilisation did not seem to be affected by sperm DNA integrity, global DNA damage, DSB and SSB were found to be correlated to embryo development outcomes. Specifically, whereas global DNA damage and DSB negatively affected (P < 0.05) the later preimplantation embryo stages (percentage of early blastocyst/blastocyst D6: for global DNA damage, R = - 0.458, and for DSB, R = - 0.551; and percentage of hatching/hatched blastocyst D6: for global DNA damage, R = - 0.505, and for DSB, R = - 0.447), global DNA damage and SSB had a negative impact (P < 0.05) on the developmental competency of fertilised embryos (R = - 0.532 and R = - 0.515, respectively). Remarkably, multiple regression analyses supported the associations found in correlation analyses. Finally, the present work also found that the inclusion of Comet assays to the conventional sperm quality tests improves the prediction of blastocyst formation (AUC = 0.9021, P < 0.05), but not fertilisation rates (P > 0.05). Conclusion: Considering all these findings, this work sets a useful model to study how SDF negatively influences fertility.
To achieve clinical impact in daily oncological practice, emerging AI-based cancer imaging research needs to have clearly defined medical focus, AI methods, and outcomes to be estimated. AI-supported cancer imaging should predict major relevant clinical endpoints, aiming to extract associations and draw inferences in a fair, robust, and trustworthy way. AI-assisted solutions as medical devices, developed using multicenter heterogeneous datasets, should be targeted to have an impact on the clinical care pathway. When designing an AI-based research study in oncologic imaging, ensuring clinical impact in AI solutions requires careful consideration of key aspects, including target population selection, sample size definition, standards, and common data elements utilization, balanced dataset splitting, appropriate validation methodology, adequate ground truth, and careful selection of clinical endpoints. Endpoints may be pathology hallmarks, disease behavior, treatment response, or patient prognosis. Ensuring ethical, safety, and privacy considerations are also mandatory before clinical validation is performed. The Artificial Intelligence for Health Imaging (AI4HI) Clinical Working Group has discussed and present in this paper some indicative Machine Learning (ML) enabled decision-support solutions currently under research in the AI4HI projects, as well as the main considerations and requirements that AI solutions should have from a clinical perspective, which can be adopted into clinical practice. If effectively designed, implemented, and validated, cancer imaging AI-supported tools will have the potential to revolutionize the field of precision medicine in oncology.
Background Ductal adenocarcinomas and neuroendocrine tumours are the two main morphological sub-types of pancreatic cancer. Using data from CONCORD-3, we examined whether the distribution of morphological sub-types could help explain international variations in pancreatic cancer survival for all morphologies combined. We also examined world-wide survival trends from pancreatic cancer, by morphological sub-type and country. Methods We estimated age-standardised one- and five-year net survival by country, calendar period of diagnosis (2000–2004, 2005–2009, 2010–2014) and morphological sub-type, using data from 295 population-based cancer registries in 58 countries for 1,258,329 adults (aged 15–99 years) diagnosed with pancreatic cancer during 2000–2014 and followed up until 31 December 2014. Results Carcinomas were by far the most common morphological sub-type, comprising 90% or more of all pancreatic tumours in all countries. Neuroendocrine tumours were rare, generally 0–10% of all tumours. During 2010–2014, age-standardised one-year net survival ranged from 10% to 30% for carcinomas, while it was much higher for neuroendocrine tumours (40% to 80%). Age-standardised five-year survival was generally poor (less than 10 %) for carcinomas, but it ranged from 20% to 50% for neuroendocrine tumours. Conclusions Survival from pancreatic carcinoma remains poor world-wide and trends showed little improvement during 2000–2014. Despite slight declines in the proportion of carcinomas, they continue to comprise the majority of pancreatic tumours. Increases in survival from neuroendocrine tumours were greater than those for carcinomas, indicating that enhancements in diagnostic techniques and treatments have helped improve survival over time.
Smart cities refer to place-specific collaborative systems where multiple actors collaborate to collectively address public problems. However, smart city initiatives regularly frame citizens as the weakest link, as passive consumers rather than active creative agents. This article argues that power imbalances between citizens and other organisational participants structurally mute citizens' voices, ultimately compromising smart cities' aims. Living laboratories are a popular smart city intervention that have the potential to address this power imbalance and empower citizens to influence smart city development. This research theoretically and empirically explores this role of living labs through a multiple-case study of urban living labs in the region of Catalonia. The findings uncover a ‘power banking’ mechanism which, coupled with other critical factors, facilitates the effectiveness of such initiatives. The considerable efforts required to engage citizens at a fairly basic level suggest that incorporating citizens into smart city models is more challenging than simplistic Quadruple Helix discourses convey.
Background Prevalence of both multimorbidity and frailty increases with age, but more evidence is needed to elucidate their relationship and their association with other health-related outcomes. We analysed the dynamics of both conditions as people age and calculate the associated risk of death, nursing home admission, and need for home care. Methods Data were drawn from the primary care electronic health records of a longitudinal cohort of people aged 65 or older in Catalonia in 2010–2019. Frailty and multimorbidity were measured using validated instruments (eFRAGICAP, a cumulative deficit model; and SNAC-K, respectively), and their longitudinal evolution was described. Cox regression models accounted for the competing risk of death and adjusted by sex, socioeconomical status, and time-varying age, alcohol and smoking. Findings We included 1 456 052 patients. Prevalence of multimorbidity was consistently high regardless of age, while frailty almost quadrupled from 65 to 99 years. Frailty worsened and also changed with age: up to 84 years, it was more related to concurrent diseases, and afterwards, to frailty-related deficits. While concurrent diseases contributed more to mortality, frailty-related deficits increased the risk of institutionalisation and the need for home care. Interpretation The nature of people’s multimorbidity and frailty vary with age, as does their impact on health status. People become frailer as they age, and their frailty is more characterised by disability and other symptoms than by diseases. Mortality is most associated with the number of comorbidities, whereas frailty-related deficits are associated with needing specialised care. Funding Instituto de Salud Carlos III through PI19/00535, and the PFIS Grant FI20/00040 (Co-funded by European Regional Development Fund/European Social Fund).
Tissue segmentation of infants could lead to early diagnosis of neurological disorders, potentially enabling early interventions. However, the challenge of tissue quantification is increased due to the very dynamic changes that happen as brain development advances over the course of the first year. One of the structural processes is the myelination which causes limited contrast between gray and white matter tissue on T1-weighted and T2-weighted magnetic resonance images at around six to nine months. In recent years, as a result of the MICCAI brain MRI segmentation challenge in 6-month old infants (iSeg17 and iSeg19), there has been an increase in interest in this complex task. In this work, we propose two methodologies to overcome issues of erroneous segmentation on the border between gray and white matter, based on knowledge-guided U-Net for segmenting the isointense infant brain. First, segmentation was guided using a prior of white matter obtained from an atlas for developing infants. Second, segmentation was focused on the low-intensity contrast boundary between white and gray matter. Experimental results on the subjects of iSeg19 challenge display the potential of utilizing the white matter prior as input for segmentation. Overall, its utilization leads to results that are closer to the brain anatomy with smoother and connected white matter regions.
Background The PRIMA phase 3 trial showed niraparib significantly prolongs median progression-free survival (PFS) versus placebo in patients with advanced ovarian cancer (OC) responsive to first-line platinum-based chemotherapy, including those who had tumors with homologous recombination deficiency (HRd). This analysis of PRIMA examined the quality-adjusted PFS (QA-PFS) and quality-adjusted time without symptoms of disease or toxicity (Q-TWiST) of patients on maintenance niraparib versus placebo. Methods Patients were randomized 2:1 to receive once-daily maintenance niraparib ( n = 487) or placebo ( n = 246). QA-PFS was defined as the PFS of patients adjusted for their health-related quality of life (HRQoL) prior to disease progression, measured using European Quality of Life Five-Dimension (EQ-5D) questionnaire index scores from the PRIMA trial. Q-TWiST was calculated by combining data on PFS, duration of symptomatic grade ⩾2 adverse events (fatigue or asthenia, nausea, vomiting, abdominal pain, and abdominal bloating) prior to disease progression, and EQ-5D index scores. Analyses used data collected up to the last date of PFS assessment (May 17, 2019). Results The restricted mean QA-PFS was significantly longer with niraparib versus placebo in the HRd ( n = 373) and overall intention-to-treat (ITT; n = 733) populations (mean gains of 6.5 [95% confidence interval; CI, 3.9–8.9] and 4.1 [95% CI, 2.2–5.8] months, respectively). There were also significant improvements in restricted mean Q-TWiST for niraparib versus placebo (mean gains of 5.9 [95% CI, 3.5–8.6] and 3.5 [95% CI, 1.7–5.6] months, respectively) in the HRd and ITT populations. Conclusions In patients with advanced OC, first-line niraparib maintenance was associated with significant gains in QA-PFS and Q-TWiST versus placebo. These findings demonstrate that niraparib maintenance treatment is associated with a PFS improvement and that treatment benefit is maintained even when HRQoL and/or toxicity data are combined with PFS in a single measure. Trial registration ClinicalTrials.gov: NCT02655016; trial registration date: January 13, 2016 Plain language summary Background: In a large clinical trial called PRIMA, patients with advanced cancer of the ovary (ovarian cancer) were given either niraparib (a type of cancer medicine) or placebo (a pill containing no medicine/active substances) after having chemotherapy (another type of cancer medicine). Taking niraparib after chemotherapy is called maintenance therapy and aims to give patients more time before their cancer returns or gets worse than if they were not given any further treatment. In the PRIMA trial, patients who took niraparib did have more time before their cancer progressed than if they took placebo. However, it is important to consider patients’ quality of life, which can be made worse by cancer symptoms and/or side effects of treatment. Here, we assessed the overall benefit of niraparib for patients in PRIMA. Methods: Both the length of time before disease progression (or survival time) and quality of life were considered using two different analyses: ● The first analysis was called quality-adjusted PFS (QA-PFS) and looked at how long patients survived with good quality of life. ● The second analysis was called quality-adjusted time without symptoms of disease or toxicity (Q-TWiST) and looked at how long patients survived without cancer symptoms or treatment side effects. Results: The PRIMA trial included 733 patients; 487 took niraparib and 246 took placebo. Around half of the patients in both groups had a type of ovarian cancer that responds particularly well to drugs like niraparib – they are known as homologous recombination deficiency (HRd) patients. ● When information on quality of life (collected from patient questionnaires) and survival was combined in the QA-PFS analysis, HRd patients who took niraparib had approximately 6.5 months longer with a good quality of life before disease progression than those who took placebo. In the overall group of patients (including HRd patients and non-HRd patients), those who took niraparib had approximately 4 months longer than with placebo. ● Using the second analysis (Q-TWiST) to combine information on survival with cancer symptoms and treatment side effects, the HRd patients taking niraparib had approximately 6 months longer without cancer symptoms or treatment side effects (such as nausea or vomiting) than patients taking placebo. In the overall group of patients, those taking niraparib had approximately 3.5 months longer without these cancer symptoms/side effects than patients receiving placebo. Conclusions: These results show that the survival benefits of niraparib treatment remain when accounting for patients’ quality of life. These benefits were seen not only in HRd patients who are known to respond better to niraparib, but in the overall group of patients who took niraparib.
Narratives on innovation in education are spreading fast and both national and local educational administrations have been recently promoting innovation policies and programmes in many different European contexts. Academic literature analysing the potential benefits of innovation in education has expanded accordingly, with some international organizations increasingly commissioning research aiming to study the impacts of educational innovation, especially on learning outcomes. Interestingly, less attention has been given to analysing how these policies and programmes are translated into different practices at the school level. Drawing on a policy enactment framework, this paper aims to analyse the ways in which schools interpret top-down policy text and prescription on innovation and enact innovation in education. To do so, we focus on the case of Xarxes per al Canvi (XC), an educational innovation programme launched by the educational administration of the city of Barcelona in 2017 that aims to create school networks in order to stimulate knowledge sharing and innovation. Findings show how schools make sense of the innovation policy diversely. Policy enactment outcomes appear to be context-sensitive, with schools enacting its precepts in different ways, especially to serve their needs in increasingly competitive local education markets.
Resumen Objetivo Determinar la prevalencia de la anemia y las características definitorias de los individuos que la padecen. Diseño Estudio observacional descriptivo de base poblacional. Variables extraídas de la base de datos informatizada de historias clínicas de atención primaria. Emplazamiento El presente estudio se realizó en la región sanitaria Metropolitana Nord, Barcelona. Participantes Pacientes con anemia evaluada mediante hemoglobina durante el año 2019. Criterios inclusión: mayores de 14 años, con determinaciones de hemoglobina por debajo del valor recomendado por la OMS. Criterio exclusión: no realizar seguimiento por la sanidad pública. Mediciones principales Variables demográficas, variables clínicas (fármacos, diagnóstico de anemia, paciente crónico complejo o enfermedad crónica avanzada. Variables consumo de recursos (número de visitas a atención primaria, número de ingresos hospitalarios y número de consultas a urgencias). Las variables demográficas y clínicas se evaluaron mediante frecuencias y porcentajes, media y desviación estándar para las variables cuantitativas. Las características de los diferentes subgrupos se compararon con la t de Student. Resultados La prevalencia de la anemia fue del 3,78%. Media de edad 64,5 años, el 60,8% mujeres. El 15,8% identificados como pacientes crónicos complejos. La hipertensión arterial presente en el 51,46%. El 50,97% fueron anemias ferropénicas. Consultaron al médico de familia 2,25 veces de media y a la enfermera de familia 1,49 veces. De las personas con criterios analíticos de anemia, solo el 46,57% tenían registrado un diagnóstico de anemia. Conclusiones En el territorio estudiado se ha objetivado un infraregistro del diagnóstico de anemia. Se observan diferencias en la caracterización por género y edad.
The implementation and use of artificial intelligence (AI), and automatic learning in particular, are ushering in a methodological and conceptual updating of creative practices and their associated theory. The roots of these changes lie in disciplinary traditions associated with research into automatism promoted in the 1960s, with the introduction of cybernetics and computational sciences. This article focuses on the spread of the notion of authorship in architecture in the digitalization process and takes the discussion of systematic automatism in home design as a case study. For doing so it revises the main theoretical references, particularly the work by Gordon Pask, unfolds projects from the 60 s to post the main questions raised by AI today in the Design Field in respect to Authorship. While Gordon Pask suggested the transformation of the architect into a systems designer, now we can state that this is an environment in which the architect is transformed into an educator and a critic, an educator in the sense that they have to design the datasets, or conditions of possibility of learning; they have to instruct the generative and discriminative neural networks through numerous groups of “examples” and guide this learning through weighted assessment protocols. KeywordsMachine learningAuthorshipDomestic space
In the present work, the effect of Si addition on the magnetic properties of Fe60−xCo25Ni15Six (x = 0, 5, 10, 20, and 30 at%) alloys prepared by mechanical alloying was analyzed by X-ray diffraction and magnetic vibrating sample magnetometry and SQUID. The crystallographic parameters of the bcc-solid solutions were calculated by Rietveld refinement of the X-ray diffraction patterns with Maud software. Scanning electron microscopy (SEM) was used to determine the morphology of the powdered alloys as a function of milling time. It was found that the Si addition has an important role in the increase of structural hardening and brittleness of the particles (favoring the more pronounced refinement of crystallites). The resulting nanostructure is highlighted in accordance with the concept of the structure of defects. Magnetic properties were related to the metalloid addition, formed phases, and chemical compositions. All processed samples showed a soft ferromagnetic behavior (Hc ≤ 100 Oe). The inhomogeneous evolution of the magnetization saturation as a function of milling time is explained by the magnetostriction effective anisotropy and stress induced during mechanical alloying.
Invasive fishes are a major environmental issue at the global scale, particularly for their impacts on freshwater ecosystems via the mechanisms of hybridisation, competition, predation, and disease transmission. This is of special conservation concern on the Iberian Peninsula due to the high level of endemism. With the aim to improve our knowledge on the invasion process of non‐native fishes, the present study consists of analysing key biological traits potentially related to their colonisation capacity: morphology, parasite communities, and genetic diversity. A non‐native population of Languedoc minnow Phoxinus septimaniae (leuciscid species native to south‐east France) was assessed in Tordera Stream (north‐eastern Iberian Peninsula). Fish were sampled in mid‐May (reproductive period) 2018 and 2019. Biological traits were compared between colonising and resident individuals from stream reaches of intermittent and perennial flow, respectively. After adjusting for fish size/body mass with analysis of covariance, body condition and health status (external and internal examination) tended to be better in the intermittent reach. Parasite load (number) and diversity (Shannon index) were significantly higher in the perennial reach. Fish development was also more stable in colonising minnow, as detected by low levels of fluctuating asymmetry for the eye diameter and the length of pectoral fins (adjusted for bilateral character size). Genetic diversity (microsatellite markers) was lower in resident minnow. Minnow genetic homozygosity (index of internal relatedness) was positively associated with parasite load and developmental instability (high fluctuating asymmetry levels). Our results show that colonising minnows display particular profiles of morphological, parasitological, and genetic traits that apparently allow these individuals to increase their dispersal capacity to survive and reproduce in harsher environmental conditions. The potential mechanisms may be related to a wider physiological tolerance, better immune response, parasite resistance/tolerance, and genetic diversity in colonising specimens. Such information reveals the specific traits of successful fish invaders under a seasonal colonisation dynamic. These data are relevant for environmental managers, to predict and identify areas vulnerable to invasion, in order to establish monitoring programmes for early detection, which will help to reduce the spread of non‐native fish populations.
Aims The aim of this study was to conduct a meta-analysis of prospective studies assessing the relationship between bundle branch block (BBB) or wide QRS and risk of all-cause mortality in patients with acute heart failure (AHF). Methods and results We searched the PubMed, Scopus and Web of Science database from inception to February 2022 to identify single centre or multicentre studies including a minimum of 400 patients and assessing the association between BBB or wide QRS and mortality in patients with AHF. Study-specific hazard ratio (HR) estimates were combined using a random-effects meta-analysis. Two meta-analyses were performed: (1) grouping by conduction disturbance and follow-up length and, (2) using the results from the longest follow-up for each study and grouping by the type of BBB. The metaanalysis included 21 publications with a total of 116,928 patients. Wide QRS (considering right (RBBB) and left (LBBB) altogether) was associated with a significant increment in the risk of all-cause mortality (pooled adjusted HR 1.112, 95% CI 1.065–1.160). The increased risk of death was also present when LBBB (HR 1.121, 95% CI 1.042–1.207) and RBBB (HR 1.187, 95% CI 1.045–1.348) were considered individually. There was no difference in risk between LBBB and RBBB (P for interaction = 0.533). Other outcomes including sudden death, rehospitalization and a combination of cardiovascular death or rehospitalization were also increased in patients with BBB or wide QRS. Conclusions This meta-analysis suggests a modest increase in the risk of all-cause mortality among patients with AHF and BBB or wide QRS, irrespective of the type of BBB.
Transient liquid assisted growth (TLAG) is an ultrafast non‐equilibrium growth process mainly governed by kinetic parameters, which are only accessible through fast in situ characterizations. In situ synchrotron X‐ray diffraction (XRD) analysis and in situ electrical resistivity measurements are used to derive kinetic diagrams of YBa2Cu3O7−x (YBCO) superconducting films prepared via TLAG and to reveal the unique peculiarities of the process. In particular, diagrams for the phase evolution and the YBCO growth rates have been built for the two TLAG routes. It is shown that TLAG transient liquids can be obtained upon the melting of two barium cuprate phases (and not just one), differentiated by their copper oxidation state. This knowledge serves as a guide to determine the processing conditions to reach high performance films at high growth rates. With proper control of these kinetic parameters, films with critical current densities of 2–2.6 MA cm⁻² at 77 K and growth rates between 100–2000 nm s⁻¹ are reached. These growth rates are 1.5–3 orders of magnitude higher than those of conventional methods.
The EU Water Framework Directive (WFD) ecological status classification of surface water bodies (SWBs) provides a synthetic assessment of how anthropogenic pressures impact on aquatic environments. In the evaluation of River Basins Management Plans of the second cycle (2015–2021), the EU commission reported the need for Spain: (i) to improve apportionment of pressures among individual sectors to be able to target appropriate measures, (ii) to take operational measures to mitigate climate change effects on the different sources of pressure affecting surface water quality status. Urban wastewater treatment plants (WWTPs) discharges to river SWBs represent an impactful point source of pressure. This study postulates that the dilution factor (DF) of WWTPs discharged to river flow can be estimated for EU Member states and can be used as suitable proxy to predict the EU SWBs ecological status. The implemented approach uses the available datasets centralized by the Environmental European Agency, and a focus on Spain’s situation is provided. Climate change effects on SWBs river flows are considered using a set of scenarios based on IPCC regionalized estimates. Results confirm that cumulative WWTP discharge and DF can be estimated and mapped for EU member states. The study confirms that it is statistically sound to use DF to assess the probability for a SWB to reach or not a good ecological status. 21% and 33.9% of the SWBs in Spain show a decreasing DF for the projections based on the high representative concentration pathway emission scenarios (RCPs 4.5; 8.5), respectively for the early (2011–2040) and midcentury (2041–2070) periods. Using the DF as an indicator, results show that in the span of these two periods, 55% of the 1046 SWBs with a current good ecological status have a higher than 70% probability to lose their good ecological status classification.KeywordsWaterbase databaseRCPsLogistic regressionWFDSurface water bodies
Background The analysis of chromatin integrity has become an important determinant of sperm quality. In frozen-thawed bovine sperm, neither the sequence of post-thaw injury events nor the dynamics of different types of sperm DNA breaks are well understood. The aim of the present work was to describe such sperm degradation aftermath focusing on DNA damage dynamics, and to assess if this parameter can predict pregnancy rates in cattle. Results A total of 75 cryopreserved ejaculates from 25 Holstein bulls were evaluated at two post-thawing periods (0-2 h and 2-4 h), analyzing global and double-stranded DNA damage through alkaline and neutral Comet assays, chromatin deprotamination and decondensation, sperm motility, viability, acrosomal status, and intracellular levels of total ROS, superoxides and calcium. Insemination of 59,605 females was conducted using sperm from the same bulls, thus obtaining the non-return to estrus rates after 90 d (NRR). Results showed an increased rate of double-stranded breaks in the first period (0-2 h: 1.29 ± 1.01%/h vs. 2-4 h: 0.13 ± 1.37%/h; P < 0.01), whereas the rate of sperm with moderate + high single-stranded breaks was higher in the second period (0-2 h: 3.52 ± 7.77 %/h vs. 2-4h: 21.06 ± 11.69 %/h; P < 0.0001). Regarding sperm physiology, viability decrease rate was different between the two periods (0-2 h: − 4.49 ± 1.79%/h vs. 2-4 h: − 2.50 ± 3.39%/h; P = 0.032), but the progressive motility decrease rate was constant throughout post-thawing incubation (0-2 h: − 4.70 ± 3.42%/h vs. 2-4 h: − 1.89 ± 2.97%/h; P > 0.05). Finally, whereas no correlations between bull fertility and any dynamic parameter were found, there were correlations between the NRR and the basal percentage of highly-damaged sperm assessed with the alkaline Comet ( Rs = − 0.563, P = 0.003), between NRR and basal progressive motility ( Rs = 0.511, P = 0.009), and between NRR and sperm with high ROS at 4 h post-thaw ( Rs = 0.564, P = 0.003). Conclusion The statistically significant correlations found between intracellular ROS, sperm viability, sperm motility, DNA damage and chromatin deprotamination suggested a sequence of events all driven by oxidative stress, where viability and motility would be affected first and sperm chromatin would be altered at a later stage, thus suggesting that bovine sperm should be used for fertilization within 2 h post-thaw. Fertility correlations supported that the assessment of global DNA damage through the Comet assay may help predict bull fertility.
Predictive monitoring of two key properties of nanocellulose, aspect ratio and yield of nanofibrillation, would help manufacturers control and optimize production processes, given the uncertainty that still surrounds their influential factors. For that, 20 different types of cellulosic and lignocellulosic micro/nanofibers produced from spruce and pine softwoods, and by different pre-treatment and fibrillation techniques, were used as training and testing datasets aiming at the development and evaluation of three machine learning models. The models used were Random Forests (RF), Linear Regression (LR) and Artificial Neural Networks (ANN), broadening the scope of our previous work (Santos et al. in Cellulose 29:5609–5622, 2022. https://doi.org/10.1007/s10570-022-04631-5 ). Performance of these models were evaluated by comparing statistical parameters such as Mean Absolute Percentage Error (MAPE) and R². For the aspect ratio and the yield of nanofibrillation, inputs were chosen among these easily controlled or measured variables: Total lignin (wt%), Cellulose (wt%), Hemicellulose (wt%), Extractives (wt%), HPH Energy Consumption (kWh/kg), Cationic Demand (µeq/g), Transmittance at 600 nm and Consistency index (Ostwald-De Waele’s k ). In both cases, the ANN models trained here provided satisfactory estimates of aspect ratio (MAPE = 4.54% and R ² = 0.96) and the yield of nanofibrillation (MAPE = 6.74% and R ² = 0.98), being able to capture the effect of the applied energy along the fibrillation process. RF and LR models resulted in correlation coefficients of 0.93 and 0.95, respectively, for aspect ratio, while for yield of nanofibrillation the correlation coefficients were 0.87 and 0.92.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.