Management of pregnancy in autoimmune rheumatic diseases: maternal disease course, gestational and neonatal outcomes and use of medications in the prospectiveItalian P-RHEUM.it study.

OBJECTIVES: To investigate pregnancy outcomes in women with autoimmune rheumatic diseases (ARD) in the Italian prospective cohort study P-RHEUM.it. METHODS: Pregnant women with different ARD were enrolled for up to 20 gestational weeks in 29 Rheumatology Centres for 5 years (2018-2023). Maternal and infant information were collected in a web-based database. RESULTS: We analysed 866 pregnancies in 851 patients (systemic lupus erythematosus was the most represented disease, 19.6%). Maternal disease flares were observed in 135 (15.6%) pregnancies. 53 (6.1%) pregnancies were induced by assisted reproduction techniques, 61 (7%) ended in miscarriage and 11 (1.3%) underwent elective termination. Obstetrical complications occurred in 261 (30.1%) pregnancies, including 2.3% pre-eclampsia. Two cases of congenital heart block were observed out of 157 pregnancies (1.3%) with anti-Ro/SSA. Regarding treatments, 244 (28.2%) pregnancies were treated with glucocorticoids, 388 (44.8%) with hydroxychloroquine, 85 (9.8%) with conventional synthetic disease-modifying anti-rheumatic drugs and 122 (14.1%) with biological disease-modifying anti-rheumatic drugs. Live births were 794 (91.7%), mostly at term (84.9%); four perinatal deaths (0.5%) occurred. Among 790 newborns, 31 (3.9%) were small-for-gestational-age and 169 (21.4%) had perinatal complications. Exclusive maternal breast feeding was received by 404 (46.7%) neonates. The Edinburgh Postnatal Depression Scale was compiled by 414 women (52.4%); 89 (21.5%) scored positive for emotional distress. CONCLUSIONS: Multiple factors including preconception counselling and treat-to-target with pregnancy-compatible medications may have contributed to mitigate disease-related risk factors, yielding limited disease flares, good pregnancy outcomes and frequency of complications which were similar to the Italian general obstetric population. Disease-specific issues need to be further addressed to plan preventative measures.

Convolutional neural networks reveal properties of reach-to-grasp encoding in posterior parietal cortex.

Deep neural networks (DNNs) are widely adopted to decode motor states from both non-invasively and invasively recorded neural signals, e.g., for realizing brain-computer interfaces. However, the neurophysiological interpretation of how DNNs make the decision based on the input neural activity is limitedly addressed, especially when applied to invasively recorded data. This reduces decoder reliability and transparency, and prevents the exploitation of decoders to better comprehend motor neural encoding. Here, we adopted an explainable artificial intelligence approach - based on a convolutional neural network and an explanation technique - to reveal spatial and temporal neural properties of reach-to-grasping from single-neuron recordings of the posterior parietal area V6A. The network was able to accurately decode 5 different grip types, and the explanation technique automatically identified the cells and temporal samples that most influenced the network prediction. Grip encoding in V6A neurons already started at movement preparation, peaking during movement execution. A difference was found within V6A: dorsal V6A neurons progressively encoded more for increasingly advanced grips, while ventral V6A neurons for increasingly rudimentary grips, with both subareas following a linear trend between the amount of grip encoding and the level of grip skills. By revealing the elements of the neural activity most relevant for each grip with no a priori assumptions, our approach supports and advances current knowledge about reach-to-grasp encoding in V6A, and it may represent a general tool able to investigate neural correlates of motor or cognitive tasks (e.g., attention and memory tasks) from single-neuron recordings.

Use of nebulized liposomal amphotericin B and posaconazole as antifungal prophylaxis in patients with severe SARS-CoV2 infection in intensive care unit.

PURPOSE: COVID-19 associated pulmonary aspergillosis (CAPA) is common and linked with high fatality rates. To assess the impact on the incidence and outcome of CAPA of an antifungal prophylaxis (AFP) we compared two cohorts of COVID-19 patients admitted to intensive care units (ICU) in Brescia, Italy, from January to August 2021. METHODS: The study cohort included all mechanically ventilated patients observed between April 2021 and August 2021 with SARS-CoV-2-pneumonia, who received AFP with oral posaconazole (200 mg every 6 h) and nebulized liposomal amphotericin B (50 mg every 2 weeks) from ICU admission to 7 days after discharge or, if applicable, until tracheostomy removal. The control cohort included COVID-19 patients admitted to the same ICU between January and March 2021 who did not receive any AFP. Subjects with CAPA at ICU admission were excluded. RESULTS: We included 270 patients, of whom 64 (23.7%) received AFP. In patients in the study group, CAPA-related mortality was significantly reduced (29% vs. 48% p = 0.04), as well as the incidence of CAPA (3.1% vs 12.1%, p = 0.03). Patients who developed CAPA were older (mean of 70-y-old vs 63-y-old, p < 0.001). One subject discontinued posaconazole due to an adverse reaction. Among the 46 patients who received it, only one patient reached an effective plasma concentration of posaconazole. CONCLUSION: AFP was associated with reduced incidence and mortality from CAPA and was well tolerated in patients with severe COVID-19. Posaconazole concentrations below the efficacy threshold in almost all patients may be attributable to drug interactions and prompt further studies to define its clinical significance.

The impact of time since SARS-Cov-2 vaccination, age, sex and comorbidities on COVID-19 outcome in hospitalized patients with SARS-CoV-2 infection.

We evaluated the impact of COVID-19 vaccination on disease outcome in hospitalized patients with SARS-CoV-2 infection with a prospective study. 745 vaccinated and 451 unvaccinated patients consecutively admitted to a COVID-19 Hospital from 1st September 2021 to 1st September 2022 were included. Compared with unvaccinated cases, vaccinated patients were older, had more comorbidities, but had a lower risk of O2 need (odds ratio, OR, 0.46; 95 % CI 0.32-0.65) by logistic regression analysis adjusted for age, sex, comorbidity and WHO COVID-19 Clinical Progression Scale at admission. The ORs for O2 need were 0.38 (0.24-0.61), 0.50 (0.30-0.83) and 0.57 (0.34-0.96) in patients vaccinated 14-120, 121-180 and > 180 days prior to hospitalization, respectively. An anti-spike Ig titer higher than 5000 U/ml was associated with a reduced risk of O2 need (OR 0.52; 95 % CI 0.30-0.92). This study shows that COVID-19 vaccination has a significant impact on COVID-19 outcomes in hospitalized patients.

rTMS over the human medial parietal cortex impairs online reaching corrections.

Indirect correlational evidence suggests that the posteromedial sector of the human parietal cortex (area hV6A) is involved in reaching corrections. We interfered with hV6A functions using repetitive transcranial magnetic stimulation (rTMS) while healthy participants performed reaching movements and in-flight adjustments of the hand trajectory in presence of unexpected target shifts. rTMS over hV6A specifically altered action reprogramming, causing deviations of the shifted trajectories, particularly along the vertical dimension (i.e., distance). This study provides evidence of the functional relevance of hV6A in action reprogramming while a sudden event requires a change in performance and shows that hV6A also plays a role in state estimation during reaching. These findings are in line with neurological data showing impairments in actions performed along the distance dimension when lesions occur in the dorsal posterior parietal cortex.

Management of periprosthetic joint infection of shoulder arthroplasty: Single-stage versus two-stage protocols. A systematic review and meta-analysis of the literature.

Background: The treatment of shoulder prosthetic joint infections (PJIs) requiring removal of the prosthesis is not well defined. This article aims to systematically review and compare the results of the literature in single-stage and two-stage protocols in the treatment of shoulder PJI. Methods: An in-depth search on PubMed/Scopus/Web of Science databases and cross-referencing search was carried out concerning the articles reporting detailed data on the topic. Results: A total of 486 shoulder PJIs were included: 137 treated with single-stage and 349 with two-stage procedure. A similar distribution between early and not-early infections (19.1% vs 80.9%) was found between the two groups. The overall rate of success in terms of PJI eradication was significantly higher in the single-stage group (95.6% vs 85.7%, p < 0.001). The non-infection-related complications rate was 13.8% in the single-stage group and 37.6% in the two-stage group (p < 0.001), the non-infection-related revision rate was 8% and 18.9%, respectively (p = 0.005). Discussion: The single-phase protocol showed a higher success rate in eradicating the infection and a lower complication rate. However, the low number of patients included, the low quality of the articles, the lack of data on clinical severity and bacteriological virulence suggest caution in conclusions.

Does Surgical Approach Influence Complication Rate of Hip Hemiarthroplasty for Femoral Neck Fractures? A Literature Review and Meta-Analysis.

Background: Femoral neck fractures are an epidemiologically significant issue with major effects on patients and health care systems, as they account for a large percentage of bone injuries in the elderly. Hip hemiarthroplasty is a common surgical procedure in the treatment of displaced femoral neck fractures. Several surgical approaches may be used to access the hip joint in case of femoral neck fractures, each with its own benefits and potential drawbacks, but none of them has consistently been found to be superior to the others. This article aims to systematically review and compare the different approaches in terms of the complication rate at the last follow-up. Methods: an in-depth search on PubMed/Scopus/Web of Science databases and a cross-referencing search was carried out concerning the articles comparing different approaches in hemiarthroplasty and reporting detailed data. Results: A total of 97,576 hips were included: 1030 treated with a direct anterior approach, 4131 with an anterolateral approach, 59,110 with a direct lateral approach, and 33,007 with a posterolateral approach. Comparing the different approaches, significant differences were found in both the overall complication rate and the rate of revision surgery performed (p < 0.05). In particular, the posterolateral approach showed a significantly higher complication rate than the lateral approach (8.4% vs. 3.2%, p < 0.001). Furthermore, the dislocation rate in the posterolateral group was significantly higher than in the other three groups considered (p < 0.026). However, the posterolateral group showed less blood loss than the anterolateral group (p < 0.001), a lower intraoperative fractures rate than the direct anterior group (p < 0.035), and shorter mean operative time than the direct lateral group (p < 0.018). Conclusions: The posterolateral approach showed a higher complication rate than direct lateral approach and a higher prosthetic dislocation rate than the other three types of surgical approaches. On the other hand, patients treated with posterolateral approach showed better outcomes in other parameters considered, such as mean operative time, mean blood loss and intraoperative fractures rate. The knowledge of the limitations of each approach and the most common associated complications can lead to choosing a surgical technique based on the patient's individual risk.

Motor decoding from the posterior parietal cortex using deep neural networks.

Objective.Motor decoding is crucial to translate the neural activity for brain-computer interfaces (BCIs) and provides information on how motor states are encoded in the brain. Deep neural networks (DNNs) are emerging as promising neural decoders. Nevertheless, it is still unclear how different DNNs perform in different motor decoding problems and scenarios, and which network could be a good candidate for invasive BCIs.Approach.Fully-connected, convolutional, and recurrent neural networks (FCNNs, CNNs, RNNs) were designed and applied to decode motor states from neurons recorded from V6A area in the posterior parietal cortex (PPC) of macaques. Three motor tasks were considered, involving reaching and reach-to-grasping (the latter under two illumination conditions). DNNs decoded nine reaching endpoints in 3D space or five grip types using a sliding window approach within the trial course. To evaluate decoders simulating a broad variety of scenarios, the performance was also analyzed while artificially reducing the number of recorded neurons and trials, and while performing transfer learning from one task to another. Finally, the accuracy time course was used to analyze V6A motor encoding.Main results.DNNs outperformed a classic Naïve Bayes classifier, and CNNs additionally outperformed XGBoost and Support Vector Machine classifiers across the motor decoding problems. CNNs resulted the top-performing DNNs when using less neurons and trials, and task-to-task transfer learning improved performance especially in the low data regime. Lastly, V6A neurons encoded reaching and reach-to-grasping properties even from action planning, with the encoding of grip properties occurring later, closer to movement execution, and appearing weaker in darkness.Significance.Results suggest that CNNs are effective candidates to realize neural decoders for invasive BCIs in humans from PPC recordings also reducing BCI calibration times (transfer learning), and that a CNN-based data-driven analysis may provide insights about the encoding properties and the functional roles of brain regions.

The influence of action on perception spans different effectors.

Perception and action are fundamental processes that characterize our life and our possibility to modify the world around us. Several pieces of evidence have shown an intimate and reciprocal interaction between perception and action, leading us to believe that these processes rely on a common set of representations. The present review focuses on one particular aspect of this interaction: the influence of action on perception from a motor effector perspective during two phases, action planning and the phase following execution of the action. The movements performed by eyes, hands, and legs have a different impact on object and space perception; studies that use different approaches and paradigms have formed an interesting general picture that demonstrates the existence of an action effect on perception, before as well as after its execution. Although the mechanisms of this effect are still being debated, different studies have demonstrated that most of the time this effect pragmatically shapes and primes perception of relevant features of the object or environment which calls for action; at other times it improves our perception through motor experience and learning. Finally, a future perspective is provided, in which we suggest that these mechanisms can be exploited to increase trust in artificial intelligence systems that are able to interact with humans.

Complementary contribution of the medial and lateral human parietal cortex to grasping: a repetitive TMS study.

The dexterous control of our grasping actions relies on the cooperative activation of many brain areas. In the parietal lobe, 2 grasp-related areas collaborate to orchestrate an accurate grasping action: dorsolateral area AIP and dorsomedial area V6A. Single-cell recordings in monkeys and fMRI studies in humans have suggested that both these areas specify grip aperture and wrist orientation, but encode these grasping parameters differently, depending on the context. To elucidate the causal role of phAIP and hV6A, we stimulated these areas, while participants were performing grasping actions (unperturbed grasping). rTMS over phAIP impaired the wrist orientation process, whereas stimulation over hV6A impaired grip aperture encoding. In a small percentage of trials, an unexpected reprogramming of grip aperture or wrist orientation was required (perturbed grasping). In these cases, rTMS over hV6A or over phAIP impaired reprogramming of both grip aperture and wrist orientation. These results represent the first direct demonstration of a different encoding of grasping parameters by 2 grasp-related parietal areas.

Anterior-posterior gradient in the integrated processing of forelimb movement direction and distance in macaque parietal cortex.

A major issue in modern neuroscience is to understand how cell populations present multiple spatial and motor features during goal-directed movements. The direction and distance (depth) of arm movements often appear to be controlled independently during behavior, but it is unknown whether they share neural resources or not. Using information theory, singular value decomposition, and dimensionality reduction methods, we compare direction and depth effects and their convergence across three parietal areas during an arm movement task. All methods show a stronger direction effect during early movement preparation, whereas depth signals prevail during movement execution. Going from anterior to posterior sectors, we report an increased number of cells processing both signals and stronger depth effects. These findings suggest a serial direction and depth processing consistent with behavioral evidence and reveal a gradient of joint versus independent control of these features in parietal cortex that supports its role in sensorimotor transformations.

New insights on single-neuron selectivity in the era of population-level approaches.

In the past, neuroscience was focused on individual neurons seen as the functional units of the nervous system, but this approach fell short over time to account for new experimental evidence, especially for what concerns associative and motor cortices. For this reason and thanks to great technological advances, a part of modern research has shifted the focus from the responses of single neurons to the activity of neural ensembles, now considered the real functional units of the system. However, on a microscale, individual neurons remain the computational components of these networks, thus the study of population dynamics cannot prescind from studying also individual neurons which represent their natural substrate. In this new framework, ideas such as the capability of single cells to encode a specific stimulus (neural selectivity) may become obsolete and need to be profoundly revised. One step in this direction was made by introducing the concept of "mixed selectivity," the capacity of single cells to integrate multiple variables in a flexible way, allowing individual neurons to participate in different networks. In this review, we outline the most important features of mixed selectivity and we also present recent works demonstrating its presence in the associative areas of the posterior parietal cortex. Finally, in discussing these findings, we present some open questions that could be addressed by future studies.

Machine learning methods detect arm movement impairments in a patient with parieto-occipital lesion using only early kinematic information.

Patients with lesions of the parieto-occipital cortex typically misreach visual targets that they correctly perceive (optic ataxia). Although optic ataxia was described more than 30 years ago, distinguishing this condition from physiological behavior using kinematic data is still far from being an achievement. Here, combining kinematic analysis with machine learning methods, we compared the reaching performance of a patient with bilateral occipitoparietal damage with that of 10 healthy controls. They performed visually guided reaches toward targets located at different depths and directions. Using the horizontal, sagittal, and vertical deviation of the trajectories, we extracted classification accuracy in discriminating the reaching performance of patient from that of controls. Specifically, accurate predictions of the patient's deviations were detected after the 20% of the movement execution in all the spatial positions tested. This classification based on initial trajectory decoding was possible for both directional and depth components of the movement, suggesting the possibility of applying this method to characterize pathological motor behavior in wider frameworks.

Synchronous Periprosthetic Joint Infections: A Scoping Review of the Literature.

Prosthetic joint infections (PJIs) occurring in multiple joints at the same time (synchronous PJI) are an extremely rare complication, frequently associated with bacteremia, and are associated with high mortality rates. The presence of three or more prosthetic joints, rheumatoid arthritis, neoplasia, bacteremia and immune-modulating therapy seem to be the recurring risk factors for synchronous PJI. In case of PJIs, all other replaced joints should be considered as potentially infected and investigated if PJI is suspected. Treatments of synchronous multiple PJIs vary and must be decided on a case-by-case basis. However, the advantages of one-stage exchange seem to outweigh the two-stage protocol, as it decreases the number of necessary surgical procedures. Nonetheless, too few studies have been conducted to allow firm conclusions about the best handling of synchronous PJI. Thus, additional studies are needed to understand this devastating complication and to design the most appropriate diagnostic and therapeutic path.

Stopping bDMARDs at the beginning of pregnancy is associated with disease flares and preterm delivery in women with rheumatoid arthritis.

Objectives: Women with Rheumatoid Arthritis (RA) can experience flares during pregnancy that might influence pregnancy outcomes. We aimed at assessing the disease course during pregnancy and identifying risk factors for flares. Methods: Data about prospectively-followed pregnancies in RA were retrospectively collected before conception, during each trimester and in the post-partum period. Clinical characteristics, disease activity (DAS28-CRP3), medication use, and pregnancy outcomes were analysed with regard to disease flares. Results: Among 73 women who had a live birth, 64 (88%) were in remission/low disease activity before conception. During pregnancy, a flare occurred in 27 (37%) patients, mainly during first and second trimester. Flares during pregnancy were associated with the discontinuation of bDMARDs at positive pregnancy test (55% of patients with flare vs. 30% of patients with no flare, p 0.034, OR 2.857, 95% CI 1.112-8.323) and a previous use of >1 bDMARDs (33% of patients with flare vs. 10% of patients with no flare, p 0.019, OR 4.1, 95%CI 1.204-13.966). Preterm pregnancies were characterised by higher values of CRP [10 mg/L (5-11) vs. 3 mg/L (2.5-5), p 0.01] and DAS28-CRP3 [4.2 (1.9-4.5) vs. 1.9 (1.7-2.6), p 0.01] during the first trimester as compared with pregnancies at term. Preterm delivery was associated with the occurrence of flare during pregnancy (flare 27% vs. no-flare 7%, p 0.034, OR 4.625, 95%CI 1.027-20.829). Conclusion: Preterm delivery in RA patients was associated with flares during pregnancy. Flares occurred more frequently after the discontinuation of bDMARDs at positive pregnancy test. Women with aggressive RA on treatment with bDMARDs should be considered as candidates for continuing bDMARDs during pregnancy in order to reduce the risk of flare and adverse pregnancy outcomes.

Are Static Spacers Superior to Articulated Spacers in the Staged Treatment of Infected Primary Knee Arthroplasty? A Systematic Review and Meta-Analysis.

In the treatment of knee periprosthetic joint infection with a two-stage protocol, static spacers allow for the local delivery of high doses of antibiotics and help to preserve soft tissue tension. Articulated spacers were introduced to better preserve flexion after the reimplantation. The aim of this systematic review is to provide a comprehensive data collection of the results of these different spacers. An in-depth search on the main clinical databases was performed concerning the studies reporting data on the topic. A total of 87 studies and 4250 spacers were included. No significant differences were found both in pooling data analysis and meta-analysis of comparative studies about infection recurrences, complications, and clinical scores. Mean active knee flexion at last follow-up after total knee reimplantation was found to be significantly higher using articulated spacers (91.6° ± 7° for static spacers vs. 100.3° ± 9.9° for articulated spacers; p < 0.001). Meta-analysis also recognized this strong significant difference (p < 0.001). This review has confirmed that articulated spacers do not appear to be inferior to static spacers regarding all clinical outcomes, while they are superior in terms of active flexion. However, the low quality of the studies and the risk for selection bias with complex patients preferentially treated with static spacers need to be accounted for.

Combined Orthoplastic Approach in Fracture-Related Infections of the Distal Tibia.

This series reports on the treatment of distal tibia (DT) fracture-related infections (FRI) with a combined orthoplastic approach. Thirteen patients were included. In eight patients with extensive bone involvement and in those with a non-healed fracture, the DT was resected ("staged approach"). In five cases, the DT was preserved ("single-stage approach"). A wide debridement was performed, and the cavity was filled with antibiotic-loaded PerOssal beads. All patients had a soft-tissue defect covered by a free vascularized flap (anterolateral thigh perforator flap in eight cases, latissimus dorsi flap in five). At the final follow-up (mean 25 months, range, 13-37), no infection recurrence was observed. In one patient, the persistence of infection was observed, and the patient underwent a repeated debridement. In two cases, a voluminous hematoma was observed. However, none of these complications impacted the final outcome. The successful treatment of FRI depends on proper debridement and obliteration of dead spaces with a flap. Therefore, when dealing with DT FRI, debridement of infected bone and soft tissues must be as radical as required, with no fear of the need for massive reconstructions.

Is Silver the New Gold? A Systematic Review of the Preclinical Evidence of Its Use in Bone Substitutes as Antiseptic.

Antibiotic-laden bone substitutes represent a viable option in the treatment of bone and joint infections with bone defects. In particular, the addition of silver ions or silver nanoparticles to bone substitutes to achieve local antiseptic activity could represent a further contribution, also helping to prevent bacterial resistance to antibiotics. An in-depth search of the main scientific databases was performed regarding the use of silver compounds for bone substitution. The available evidence is still limited to the preclinical level: 22 laboratory studies, 2 animal models, and 3 studies, with both in vitro and in vivo analysis, were found on the topic. Numerous biomaterials have been evaluated. In vitro studies confirmed that silver in bone substitutes retains the antibacterial activity already demonstrated in coatings materials. Cytotoxicity was generally found to be low and only related to silver concentrations higher than those sufficient to achieve antibacterial activity. Instead, there are only a few in vivo studies, which appear to confirm antibacterial efficacy, although there is insufficient evidence on the pharmacokinetics and safety profile of the compounds investigated. In conclusion, research on bone substitutes doped with silver is in its early stages, but the preliminary findings seem promising.

Follicular helper T cell signature of replicative exhaustion, apoptosis, and senescence in common variable immunodeficiency.

Common variable immunodeficiency (CVID) is the most frequent primary antibody deficiency whereby follicular helper T (Tfh) cells fail to establish productive responses with B cells in germinal centers. Here, we analyzed the frequency, phenotype, transcriptome, and function of circulating Tfh (cTfh) cells in CVID patients displaying autoimmunity as an additional phenotype. A group of patients showed a high frequency of cTfh1 cells and a prominent expression of PD-1 and ICOS as well as a cTfh mRNA signature consistent with highly activated, but exhausted, senescent, and apoptotic cells. Plasmatic CXCL13 levels were elevated in this group and positively correlated with cTfh1 cell frequency and PD-1 levels. Monoallelic variants in RTEL1, a telomere length- and DNA repair-related gene, were identified in four patients belonging to this group. Their blood lymphocytes showed shortened telomeres, while their cTfh were more prone to apoptosis. These data point toward a novel pathogenetic mechanism in CVID, whereby alterations in DNA repair and telomere elongation might predispose to antibody deficiency. A Th1, highly activated but exhausted and apoptotic cTfh phenotype was associated with this form of CVID.

Decoding sensorimotor information from superior parietal lobule of macaque via Convolutional Neural Networks.

Despite the well-recognized role of the posterior parietal cortex (PPC) in processing sensory information to guide action, the differential encoding properties of this dynamic processing, as operated by different PPC brain areas, are scarcely known. Within the monkey's PPC, the superior parietal lobule hosts areas V6A, PEc, and PE included in the dorso-medial visual stream that is specialized in planning and guiding reaching movements. Here, a Convolutional Neural Network (CNN) approach is used to investigate how the information is processed in these areas. We trained two macaque monkeys to perform a delayed reaching task towards 9 positions (distributed on 3 different depth and direction levels) in the 3D peripersonal space. The activity of single cells was recorded from V6A, PEc, PE and fed to convolutional neural networks that were designed and trained to exploit the temporal structure of neuronal activation patterns, to decode the target positions reached by the monkey. Bayesian Optimization was used to define the main CNN hyper-parameters. In addition to discrete positions in space, we used the same network architecture to decode plausible reaching trajectories. We found that data from the most caudal V6A and PEc areas outperformed PE area in the spatial position decoding. In all areas, decoding accuracies started to increase at the time the target to reach was instructed to the monkey, and reached a plateau at movement onset. The results support a dynamic encoding of the different phases and properties of the reaching movement differentially distributed over a network of interconnected areas. This study highlights the usefulness of neurons' firing rate decoding via CNNs to improve our understanding of how sensorimotor information is encoded in PPC to perform reaching movements. The obtained results may have implications in the perspective of novel neuroprosthetic devices based on the decoding of these rich signals for faithfully carrying out patient's intentions.