Automated Generation of Clinical Reports Using Sensing Technologies with Deep Learning Techniques.

This study presents a pioneering approach that leverages advanced sensing technologies and data processing techniques to enhance the process of clinical documentation generation during medical consultations. By employing sophisticated sensors to capture and interpret various cues such as speech patterns, intonations, or pauses, the system aims to accurately perceive and understand patient-doctor interactions in real time. This sensing capability allows for the automation of transcription and summarization tasks, facilitating the creation of concise and informative clinical documents. Through the integration of automatic speech recognition sensors, spoken dialogue is seamlessly converted into text, enabling efficient data capture. Additionally, deep models such as Transformer models are utilized to extract and analyze crucial information from the dialogue, ensuring that the generated summaries encapsulate the essence of the consultations accurately. Despite encountering challenges during development, experimentation with these sensing technologies has yielded promising results. The system achieved a maximum ROUGE-1 metric score of 0.57, demonstrating its effectiveness in summarizing complex medical discussions. This sensor-based approach aims to alleviate the administrative burden on healthcare professionals by automating documentation tasks and safeguarding important patient information. Ultimately, by enhancing the efficiency and reliability of clinical documentation, this innovative method contributes to improving overall healthcare outcomes.

A Comprehensive Study on Pain Assessment from Multimodal Sensor Data.

Pain assessment is a critical aspect of healthcare, influencing timely interventions and patient well-being. Traditional pain evaluation methods often rely on subjective patient reports, leading to inaccuracies and disparities in treatment, especially for patients who present difficulties to communicate due to cognitive impairments. Our contributions are three-fold. Firstly, we analyze the correlations of the data extracted from biomedical sensors. Then, we use state-of-the-art computer vision techniques to analyze videos focusing on the facial expressions of the patients, both per-frame and using the temporal context. We compare them and provide a baseline for pain assessment methods using two popular benchmarks: UNBC-McMaster Shoulder Pain Expression Archive Database and BioVid Heat Pain Database. We achieved an accuracy of over 96% and over 94% for the F1 Score, recall and precision metrics in pain estimation using single frames with the UNBC-McMaster dataset, employing state-of-the-art computer vision techniques such as Transformer-based architectures for vision tasks. In addition, from the conclusions drawn from the study, future lines of work in this area are discussed.

Shot Boundary Detection with 3D Depthwise Convolutions and Visual Attention.

Shot boundary detection is the process of identifying and locating the boundaries between individual shots in a video sequence. A shot is a continuous sequence of frames that are captured by a single camera, without any cuts or edits. Recent investigations have shown the effectiveness of the use of 3D convolutional networks to solve this task due to its high capacity to extract spatiotemporal features of the video and determine in which frame a transition or shot change occurs. When this task is used as part of a scene segmentation use case with the aim of improving the experience of viewing content from streaming platforms, the speed of segmentation is very important for live and near-live use cases such as start-over. The problem with models based on 3D convolutions is the large number of parameters that they entail. Standard 3D convolutions impose much higher CPU and memory requirements than do the same 2D operations. In this paper, we rely on depthwise separable convolutions to address the problem but with a scheme that significantly reduces the number of parameters. To compensate for the slight loss of performance, we analyze and propose the use of visual self-attention as a mechanism of improvement.

The interaction between liver cirrhosis, infection by Streptococcus bovis, and colon cancer.

Whether cirrhotic patients with Streptococcus bovis bacteremia have an increased risk of colorectal neoplasm is uncertain. A multicentric retrospective cohort study was conducted investigating associations between S. bovis biotype and species, cirrhosis, and colorectal neoplasm. Out of 779 patients with S. bovis bacteremia, 69 (8.7%) had cirrhosis. No differences were found in the prevalence of colorectal neoplasm between cirrhotic and non-cirrhotic patients undergoing colonoscopy. Among cirrhotic patients, prevalence of colorectal neoplasms was higher in S. bovis biotype I (S. gallolyticus) bacteremia (80%) than in S. bovis biotype II (33.3%; p < 0.007). In conclusion, risk of colorectal neoplasm is high among cirrhotic patients with S. gallolyticus bacteremia.

POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments.

Certain fields present significant challenges when attempting to train complex Deep Learning architectures, particularly when the available datasets are limited and imbalanced. Real-time object detection in maritime environments using aerial images is a notable example. Although SeaDronesSee is the most extensive and complete dataset for this task, it suffers from significant class imbalance. To address this issue, we present POSEIDON, a data augmentation tool specifically designed for object detection datasets. Our approach generates new training samples by combining objects and samples from the original training set while utilizing the image metadata to make informed decisions. We evaluate our method using YOLOv5 and YOLOv8 and demonstrate its superiority over other balancing techniques, such as error weighting, by an overall improvement of 2.33% and 4.6%, respectively.

Species and biotypes of Streptococcus bovis causing infective endocarditis / Especies y biotipos de Streptococcus bovis causantes de endocarditis infecciosa

Introduction: Streptococcus bovis/equinus complex (SBEC) is a major cause of infective endocarditis (IE), although its incidence varies greatly depending on the geographical area. The characteristics of IE caused by Streptococcus gallolyticus susp. gallolyticus are well known; there are hardly any descriptions of IE caused by other species or biotypes. Methods: Retrospective cohort study, from 1990 to 2019, of all SBEC IE in adults in three Spanish hospitals, Lugo (LH), Barcelona (BH) and Ferrol (FH) where the population is mainly rural, urban and mixed, respectively. The incidence of IE was analyzed in 3 areas. Clinical characteristics of IE (277 cases, 258 biotyped) were compared according to SBEC species and biotypes. Results: There are significant differences between the incidence of SBEC IE in HL (27.9/106) vs. HF and HB (8.8 and 7,1, respectively, p<0.001). We found significant differences (SbI vs. SbII) in mean age (68.5 vs. 73 years; p<0.01), duration of symptoms before diagnosis (46.9±46.5 vs. 30.4±40.9 days; p<0.01), presence of comorbidities: 39.1% (78) vs. 54.2% (32; p<0.04), predisposing heart illness:62.3% (124) vs. 81.3% (48; p<0.006), particularly, prosthetic or intravascular devices IE: 24.6% (49) vs. 52.4% (31; p<0.001), bi-valve involvement:23.6% (47) vs. 11.8% (7; p<0.05) and heart failure: 24.6% (49) vs. 38.9% (23; p<0.03). There were no significant differences in embolic events, need for surgery or mortality. The association with CRC was high in both groups: 77.7% vs. 66.6%. Conclusions: IE due to SBEC has geographical variations in incidence and different clinical characteristics among biotypes. The association with CRC was high.(AU)

Introducción: El complejo Streptococcus bovis/equinus (SBEC) es una de las principales causas de endocarditis infecciosa (EI), aunque su incidencia es muy variable según la zona geográfica. Las características de EI causada por Streptococcus gallolyticus subsp. gallolyticus son bien conocidas; apenas hay descripciones de EI causada por otras especies o biotipos. Métodos: Estudio de cohorte retrospectivo, desde 1990 hasta 2019, de todas las EI por SBEC en adultos en 3 hospitales españoles, Lugo (LH), Barcelona (BH) y Ferrol (FH) donde la población es mayoritariamente rural, urbana y mixta, respectivamente. Se analizó la incidencia de EI en 3 áreas. Se compararon las características clínicas de EI (277 casos, 258 biotipados) según las especies y biotipos de SBEC. Resultados: Existen diferencias significativas entre la incidencia de EI por SBEC en HL (27,9/106) vs. HF y HB (8,8 y 7,1, respectivamente, p<0,001). Encontramos diferencias significativas (SbI vs. SbII) en edad media (68,5 vs. 73 años; p<0,01), duración de los síntomas antes del diagnóstico (46,9±46,5 vs. 30,4±40,9 días; p<0,01); comorbilidades: 39,1 (78) vs. 54,2% (32; p<0,04); enfermedad cardíaca predisponente: 62,3 (124) vs. 81,3% (48; p<0,006), en particular, EI protésica o sobre dispositivos intravasculares: 24,6 (49) vs. 52,4% (31; p<0,001); afectación bivalva: 23,6 (47) vs. 11,8% (7; p<0,05) e insuficiencia cardiaca: 24,6 (49) vs. 38,9% (23; p<0,03). No hubo diferencias significativas en cuanto a eventos embólicos, necesidad de cirugía o mortalidad. La asociación con el CCR fue alta en ambos grupos: 77,7 vs. 66,6%. Conclusiones: La EI por SBEC tiene variaciones geográficas en la incidencia y diferentes características clínicas entre los biotipos. La asociación con el CCR fue elevada.(AU)

Design, Implementation, and Empirical Validation of a Framework for Remote Car Driving Using a Commercial Mobile Network.

Despite the fact that autonomous driving systems are progressing in terms of their automation levels, the achievement of fully self-driving cars is still far from realization. Currently, most new cars accord with the Society of Automotive Engineers (SAE) Level 2 of automation, which requires the driver to be able to take control of the car when needed: for this reason, it is believed that between now and the achievement of fully automated self-driving car systems, there will be a transition, in which remote driving cars will be a reality. In addition, there are tele-operation-use cases that require remote driving for health or safety reasons. However, there is a lack of detailed design and implementation available in the public domain for remote driving cars: therefore, in this work we propose a functional framework for remote driving vehicles. We implemented a prototype, using a commercial car. The prototype was connected to a commercial 4G/5G mobile network, and empirical experiments were conducted, to validate the prototype's functions, and to evaluate its performance in real-world driving conditions. The design, implementation, and empirical evaluation provided detailed technical insights into this important research and innovation area.

Species and biotypes of Streptococcus bovis causing infective endocarditis.

INTRODUCTION: Streptococcus bovis/equinus complex (SBEC) is a major cause of infective endocarditis (IE), although its incidence varies greatly depending on the geographical area. The characteristics of IE caused by Streptococcus gallolyticus susp. gallolyticus are well known; there are hardly any descriptions of IE caused by other species or biotypes. METHODS: Retrospective cohort study, from 1990 to 2019, of all SBEC IE in adults in three Spanish hospitals, Lugo (LH), Barcelona (BH) and Ferrol (FH) where the population is mainly rural, urban and mixed, respectively. The incidence of IE was analyzed in 3 areas. Clinical characteristics of IE (277 cases, 258 biotyped) were compared according to SBEC species and biotypes. RESULTS: There are significant differences between the incidence of SBEC IE in HL (27.9/106) vs. HF and HB (8.8 and 7,1, respectively, p<0.001). We found significant differences (SbI vs. SbII) in mean age (68.5 vs. 73 years; p<0.01), duration of symptoms before diagnosis (46.9±46.5 vs. 30.4±40.9 days; p<0.01), presence of comorbidities: 39.1% (78) vs. 54.2% (32; p<0.04), predisposing heart illness:62.3% (124) vs. 81.3% (48; p<0.006), particularly, prosthetic or intravascular devices IE: 24.6% (49) vs. 52.4% (31; p<0.001), bi-valve involvement:23.6% (47) vs. 11.8% (7; p<0.05) and heart failure: 24.6% (49) vs. 38.9% (23; p<0.03). There were no significant differences in embolic events, need for surgery or mortality. The association with CRC was high in both groups: 77.7% vs. 66.6%. CONCLUSIONS: IE due to SBEC has geographical variations in incidence and different clinical characteristics among biotypes. The association with CRC was high.

A Review on Deep Learning Techniques for Video Prediction.

The ability to predict, anticipate and reason about future outcomes is a key component of intelligent decision-making systems. In light of the success of deep learning in computer vision, deep-learning-based video prediction emerged as a promising research direction. Defined as a self-supervised learning task, video prediction represents a suitable framework for representation learning, as it demonstrated potential capabilities for extracting meaningful representations of the underlying patterns in natural videos. Motivated by the increasing interest in this task, we provide a review on the deep learning methods for prediction in video sequences. We first define the video prediction fundamentals, as well as mandatory background concepts and the most used datasets. Next, we carefully analyze existing video prediction models organized according to a proposed taxonomy, highlighting their contributions and their significance in the field. The summary of the datasets and methods is accompanied with experimental results that facilitate the assessment of the state of the art on a quantitative basis. The paper is summarized by drawing some general conclusions, identifying open research challenges and by pointing out future research directions.

A Survey of Alzheimer's Disease Early Diagnosis Methods for Cognitive Assessment.

Dementia is a syndrome that is characterised by the decline of different cognitive abilities. A high rate of deaths and high cost for detection, treatments, and patients care count amongst its consequences. Although there is no cure for dementia, a timely diagnosis helps in obtaining necessary support, appropriate medication, and maintenance, as far as possible, of engagement in intellectual, social, and physical activities. The early detection of Alzheimer Disease (AD) is considered to be of high importance for improving the quality of life of patients and their families. In particular, Virtual Reality (VR) is an expanding tool that can be used in order to assess cognitive abilities while navigating through a Virtual Environment (VE). The paper summarises common AD screening and diagnosis techniques focusing on the latest approaches that are based on Virtual Environments, behaviour analysis, and emotions recognition, aiming to provide more reliable and non-invasive diagnostics at home or in a clinical environment. Furthermore, different AD diagnosis evaluation methods and metrics are presented and discussed together with an overview of the different datasets.

Long-Term Carbapenems Antimicrobial Stewardship Program.

OBJECTIVE: To evaluate clinical and antibiotic resistance impact of carbapenems stewardship programs. METHODS: descriptive study, pre-post-intervention, between January 2012 and December 2019; 350-bed teaching hospital. Prospective audit and feedback to prescribers was carried out between January 2015 and December 2019. We evaluate adequacy of carbapenems prescription to local guidelines and compare results between cases with accepted or rejected intervention. Analysis of antibiotic-consumption and hospital-acquired multidrug-resistant (MDR) bloodstream infections (BSIs) was performed. RESULTS: 1432 patients were followed. Adequacy of carbapenems prescription improved from 49.7% in 2015 to 80.9% in 2019 (p < 0.001). Interventions on prescription were performed in 448 (31.3%) patients without carbapenem-justified treatment, in 371 intervention was accepted, in 77 it was not. Intervention acceptance was associated with shorter duration of all antibiotic treatment and inpatient days (p < 0.05), without differences in outcome. During the period 2015-2019, compared with 2012-2014, decreased meropenem consumption (Rate Ratio 0.58; 95%CI: 0.55-0.63), candidemia and hospital-acquired MDR BSIs rate (RR 0.62; 95%CI: 0.41-0.92, p = 0.02), and increased cefepime (RR 2; 95%CI: 1.77-2.26) and piperacillin-tazobactam consumption (RR 1.17; 95%CI: 1.11-1.24), p < 0.001. CONCLUSIONS: the decrease and better use of carbapenems achieved could have clinical and ecological impact over five years, reduce inpatient days, hospital-acquired MDR BSIs, and candidemia, despite the increase in other antibiotic-consumption.

Evaluation of different chrominance models in the detection and reconstruction of faces and hands using the growing neural gas network.

Physical traits such as the shape of the hand and face can be used for human recognition and identification in video surveillance systems and in biometric authentication smart card systems, as well as in personal health care. However, the accuracy of such systems suffers from illumination changes, unpredictability, and variability in appearance (e.g. occluded faces or hands, cluttered backgrounds, etc.). This work evaluates different statistical and chrominance models in different environments with increasingly cluttered backgrounds where changes in lighting are common and with no occlusions applied, in order to get a reliable neural network reconstruction of faces and hands, without taking into account the structural and temporal kinematics of the hands. First a statistical model is used for skin colour segmentation to roughly locate hands and faces. Then a neural network is used to reconstruct in 3D the hands and faces. For the filtering and the reconstruction we have used the growing neural gas algorithm which can preserve the topology of an object without restarting the learning process. Experiments conducted on our own database but also on four benchmark databases (Stirling's, Alicante, Essex, and Stegmann's) and on deaf individuals from normal 2D videos are freely available on the BSL signbank dataset. Results demonstrate the validity of our system to solve problems of face and hand segmentation and reconstruction under different environmental conditions.

New tools and approaches to promote soil and water bioengineering in the Mediterranean.

Soil loss and erosion is a major environmental problem in the Mediterranean. Soil and water bioengineering uses plants and/or parts of plants along with inert material to create solutions to fulfill soil conservation objectives combined with an ecological rehabilitation approach. The ECOMED project developed novel approaches and tools to specialize the soil and water bioengineering sector within the Mediterranean. The first activity was the Sector Needs Analysis were the responses to an online questionnaire of 110 stakeholders from the region were analyzed. The main conclusion was the need to specialize the soil and water bioengineering sector in the Mediterranean. In addition, 21 soil and water bioengineering case studies in the Mediterranean were reviewed. Many works of this type are implemented in the region, but have flaws because of the lack of training material, design routines, protocols, specific to the region. The second activity developed New Design Routines and Protocols. Specifically, three protocols, one template and one plant database were developed for the region. Finally, in the Training Material activity, six educational modules along with a handbook (that contained modules, protocols, template and case studies) were developed. Overall the ECOMED project generated new and novel material and tools that were lacking in the region to enhance the specialization process of the soil and water bioengineering sector. These should increase the adoption of soil and water bioengineering techniques with better trained and new professionals as well as improve the work performance of these interventions.

Long-term efficacy of 6-month therapy with isoniazid and rifampin compared with isoniazid, rifampin, and pyrazinamide treatment for pleural tuberculosis.

Research into anti-tuberculosis treatment has mainly focused on pulmonary tuberculosis (TB), with few studies on pleural-TB. The aim of the study is to compare the long-term efficacy of a 6-month treatment regimen with isoniazid and rifampicin (6HR) with treatment regimen of isoniazid, rifampicin, and pyrazinamide (6HR2Z) for pleural-TB. A case-control study of 200 HIV-negative patients with pleural-TB prospectively followed in our TB-unit from 1995 to 2018. The primary resistance to isoniazid is < 4% in our geographic area. Pleural-TB diagnosis was based on a positive culture for M. tuberculosis (84 patients), presence of caseating granulomas in pleural biopsy (28), or characteristics of pleural fluid (88). A comparative study of demographic and clinical characteristics between the treatment groups was carried out. Out of the 200 patients followed, (112 males, 88 females; mean age 32.9 ± 18.4 years), 99 patients were treated with 6HR regimen and 101 with 6HR2Z. The groups were comparable, except the 6HR2Z had larger size of pleural effusion. All patients completed the treatment. The group treated with 6HR presented fewer adverse effects (15.3%) than 6HR2Z group (33%), p = 0.005, and lower frequency of severe hepatic toxicity (5% vs 10.9%). Four patients died from causes other than TB during treatment with 6HR2Z, and all other patients were cured during a monitoring period for 8.4 years (IQRs, 3.3-14.3). Six patients in 6HR and 10 in 6HR2Z developed residual pachypleuritis. 6HR is as effective as 6HR2Z treatment for pleural-TB, with fewer adverse effects.

Modificaciones en la capa de fibras nerviosas poscirugía de catarata en glaucoma primario de ángulo abierto / Changes undergone by the nerve fiber layer after cataract surgery in patients with primary open angle glaucoma

RESUMEN Objetivo: Evaluar la modificación de la capa de fibras nerviosas después de la cirugía de catarata en pacientes con glaucoma primario de ángulo abierto. Métodos: Se realizó un estudio prexperimental en dos grupos de pacientes, 31 de ellos con glaucoma primario de ángulo abierto (49 ojos) y 29 pacientes sin glaucoma (50 ojos). Se analizó el comportamiento del espesor de la capa de fibras nerviosas retinianas antes y después de la cirugía de catarata (facoemulsificación). Fue evaluado el comportamiento pre y posoperatorio de la agudeza visual, así como la presión intraocular. Resultados: La edad promedio fue 71,8 ± 8,0 años en el grupo sin glaucoma y 67,8 ± 8,9 años en el grupo con glaucoma primario de ángulo abierto. La mediana de la agudeza visual sin corrección pasó de 0,20 a 0,90 en el grupo sin glaucoma y 0,30 a 0,80 en el grupo con glaucoma primario de ángulo abierto, donde resultó la p< 0,05 en ambos casos. La mediana de la presión intraocular cayó 1 mmHg en ambos grupos. En los cuadrantes superior e inferior hubo un incremento del espesor de la capa de fibras después de la operación en ambos grupos, pero solo se observó en cuadrantes nasal y temporal en el grupo con glaucoma. Conclusiones: Después de la cirugía de catarata se produce un incremento del espesor de la capa de fibras. La opacidad impide la transmisión de la señal a la retina(AU)

ABSTRACT Objective: Evaluate the changes undergone by the nerve fiber layer after cataract surgery in patients with primary open angle glaucoma. Methods: A pre-experimental study was conducted of two groups of patients, 31 with primary open angle glaucoma (49 eyes) and 29 without glaucoma (50 eyes). The behavior of retinal nerve fiber layer thickness was analyzed before and after cataract surgery (phacoemulsification). Visual acuity and intraocular pressure were evaluated during the pre- and postoperative periods. Results: Mean age was 71.8 ± 8.0 years in the group without glaucoma and 67.8 ± 8.9 years in the group with primary open angle glaucoma. Median visual acuity without correction rose from 0.20 to 0.90 in the group without glaucoma, and from 0.30 to 0.80 in the group with primary open angle glaucoma, with p < 0.05 in both. Median intraocular pressure fell 1 mmHg in both groups. An increase in fiber layer thickness was observed after surgery in the upper and lower quadrants of both groups, but only in the nasal and temporal quadrants of the group with glaucoma. Conclusions: Fiber layer thickness increased after cataract surgery. Opacity hindered signal transmission to the retina(AU)

Calidad de vida relativa a la función visual en pacientes operados de catarata / Vision-related quality of life of patients undergoing cataract surgery

RESUMEN Objetivo: Determinar el comportamiento de la calidad de vida relativa a la función visual en pacientes operados de catarata en el Instituto Cubano de Oftalmología Ramón Pando Ferrer en el periodo enero a julio del año 2017. Métodos: Se realizó un estudio descriptivo, prospectivo y longitudinal. Para esto fueron reclutados de manera consecutiva todos los pacientes que fueron tributarios de cirugía de catarata, que cumplieron con los criterios de inclusión, atendidos en el periodo señalado. A estos se les aplicó el cuestionario NEI-VFQ-23 antes de la cirugía y al mes de realizada la intervención quirúrgica. Los resultados del cuestionario se correlacionaron con la edad, la agudeza visual y la presencia de complicaciones. Resultados: Se estudiaron 205 pacientes (410 ojos) con una edad promedio de 69,7 años. El 55,1 por ciento de los pacientes era del sexo femenino; el 67,3 por ciento presentaba algún tipo de co-morbilidad sistémica y el 23,4 por ciento presentó algún antecedente de enfermedad y/o cirugía ocular. La agudeza visual mejoró en el posoperatorio. La calidad de vida relativa a la función visual es notablemente superior después de la cirugía de catarata, y estos resultados varían según la edad, la presencia de comorbilidades sistémicas y los antecedentes oculares. Este estudio es clínico y estadísticamente significativo. Conclusiones: La cirugía de catarata mejora significativamente la calidad de vida relativa a la función visual de los pacientes(AU)

ABSTRACT Objective: Determine the vision-related quality of life of patients undergoing cataract surgery at Ramón Pando Ferrer Cuban Institute of Ophthalmology from January to July 2017. Methods: A prospective longitudinal descriptive study was conducted based on consecutive recruitment of all the patients undergoing cataract surgery in the study period who met the inclusion criteria. The NEI-VFQ-23 questionnaire was applied to all participant patients before the operation and one month after surgery. Results of the questionnaire were correlated with age, visual acuity and the presence of complications. Results: The study sample was 205 patients (410 eyes); mean age was 69.7 years. 55.1 percent of the patients were female; 67.3 percent presented some sort of systemic comorbidity; and 23,4 percent had a history of disease and/or eye surgery. Visual acuity improved during the postoperative period. Vision-related quality of life is notably better after cataract surgery. These results vary according to age, the presence of systemic comorbidities and ocular antecedents. The study is clinically and statistically significant. Conclusions: Cataract surgery significantly improves the vision-related quality of life of patients(AU)

Compressed kNN: K-Nearest Neighbors with Data Compression.

The kNN (k-nearest neighbors) classification algorithm is one of the most widely used non-parametric classification methods, however it is limited due to memory consumption related to the size of the dataset, which makes them impractical to apply to large volumes of data. Variations of this method have been proposed, such as condensed KNN which divides the training dataset into clusters to be classified, other variations reduce the input dataset in order to apply the algorithm. This paper presents a variation of the kNN algorithm, of the type structure less NN, to work with categorical data. Categorical data, due to their nature, can be compressed in order to decrease the memory requirements at the time of executing the classification. The method proposes a previous phase of compression of the data to then apply the algorithm on the compressed data. This allows us to maintain the whole dataset in memory which leads to a considerable reduction of the amount of memory required. Experiments and tests carried out on known datasets show the reduction in the volume of information stored in memory and maintain the accuracy of the classification. They also show a slight decrease in processing time because the information is decompressed in real time (on-the-fly) while the algorithm is running.