Coupling between the prelimbic cortex, nucleus reuniens, and hippocampus during NREM sleep remains stable under cognitive and homeostatic demands.

The interplay between the medial prefrontal cortex and hippocampus during non-rapid eye movement (NREM) sleep contributes to the consolidation of contextual memories. To assess the role of the thalamic nucleus reuniens (Nre) in this interaction, we investigated the coupling of neuro-oscillatory activities among prelimbic cortex, Nre, and hippocampus across sleep states and their role in the consolidation of contextual memories using multi-site electrophysiological recordings and optogenetic manipulations. We showed that ripples are time-locked to the Up state of cortical slow waves, the transition from UP to DOWN state in thalamic slow waves, the troughs of cortical spindles, and the peaks of thalamic spindles during spontaneous sleep, rebound sleep and sleep following a fear conditioning task. In addition, spiking activity in Nre increased before hippocampal ripples, and the phase-locking of hippocampal ripples and thalamic spindles during NREM sleep was stronger after acquisition of a fear memory. We showed that optogenetic inhibition of Nre neurons reduced phase-locking of ripples to cortical slow waves in the ventral hippocampus whilst their activation altered the preferred phase of ripples to slow waves in ventral and dorsal hippocampi. However, none of these optogenetic manipulations of Nre during sleep after acquisition of fear conditioning did alter sleep-dependent memory consolidation. Collectively, these results showed that Nre is central in modulating hippocampus and cortical rhythms during NREM sleep.

Negation recognition in clinical natural language processing using a combination of the NegEx algorithm and a convolutional neural network.

BACKGROUND: Important clinical information of patients is present in unstructured free-text fields of Electronic Health Records (EHRs). While this information can be extracted using clinical Natural Language Processing (cNLP), the recognition of negation modifiers represents an important challenge. A wide range of cNLP applications have been developed to detect the negation of medical entities in clinical free-text, however, effective solutions for languages other than English are scarce. This study aimed at developing a solution for negation recognition in Spanish EHRs based on a combination of a customized rule-based NegEx layer and a convolutional neural network (CNN). METHODS: Based on our previous experience in real world evidence (RWE) studies using information embedded in EHRs, negation recognition was simplified into a binary problem ('affirmative' vs. 'non-affirmative' class). For the NegEx layer, negation rules were obtained from a publicly available Spanish corpus and enriched with custom ones, whereby the CNN binary classifier was trained on EHRs annotated for clinical named entities (cNEs) and negation markers by medical doctors. RESULTS: The proposed negation recognition pipeline obtained precision, recall, and F1-score of 0.93, 0.94, and 0.94 for the 'affirmative' class, and 0.86, 0.84, and 0.85 for the 'non-affirmative' class, respectively. To validate the generalization capabilities of our methodology, we applied the negation recognition pipeline on EHRs (6,710 cNEs) from a different data source distribution than the training corpus and obtained consistent performance metrics for the 'affirmative' and 'non-affirmative' class (0.95, 0.97, and 0.96; and 0.90, 0.83, and 0.86 for precision, recall, and F1-score, respectively). Lastly, we evaluated the pipeline against two publicly available Spanish negation corpora, the IULA and NUBes, obtaining state-of-the-art metrics (1.00, 0.99, and 0.99; and 1.00, 0.93, and 0.96 for precision, recall, and F1-score, respectively). CONCLUSION: Negation recognition is a source of low precision in the retrieval of cNEs from EHRs' free-text. Combining a customized rule-based NegEx layer with a CNN binary classifier outperformed many other current approaches. RWE studies highly benefit from the correct recognition of negation as it reduces false positive detections of cNE which otherwise would undoubtedly reduce the credibility of cNLP systems.

Sleep as a window on the sensorimotor foundations of the developing hippocampus.

The hippocampal formation plays established roles in learning, memory, and related cognitive functions. Recent findings also suggest that the hippocampus integrates sensory feedback from self-generated movements to modulate ongoing motor responses in a changing environment. Such findings support the view of Bland and Oddie (Behavioural Brain Research, 2001, 127, 119-136) that the hippocampus is a site of sensorimotor integration. In further support of this view, we review neurophysiological evidence in developing rats that hippocampal function is built on a sensorimotor foundation and that this foundation is especially evident early in development. Moreover, at those ages when the hippocampus is first establishing functional connectivity with distant sensory and motor structures, that connectivity is preferentially expressed during periods of active (or REM) sleep. These findings reinforce the notion that sleep, as the predominant state of early infancy, provides a critical context for sensorimotor development, including development of the hippocampus and its associated network.

Single-shot optical surface profiling using extended depth of field 3D microscopy.

The measurement of three-dimensional samples at high speed is essential for many applications, either due to the requirement for measuring samples that change fast over time, or due to the requirement of reducing the scanning time, and therefore inspection cost, in industrial environments. Conventionally, the measurement of surface topographies at high resolution typically requires an axial scanning of the sample. We report the implementation of a technique able to reconstruct surface topographies at high resolution, only from the acquisition of a single camera shot, dropping the need to perform an axial scan. A system prototype is reported and assessed as an ultra-fast optical surface profiler. We propose robust calibration and operation methods and algorithms to reconstruct surface topographies of optically-rough samples, and compare the experimental results with a reference confocal optical profiler.

The impact of COVID-19 on patients with asthma.

BACKGROUND: An association between the severity of coronavirus disease 2019 (COVID-19) and the presence of certain chronic conditions has been suggested. However, unlike influenza and other viruses, the disease burden of COVID-19 in patients with asthma has been less evident. OBJECTIVE: To understand the impact of COVID-19 in patients with asthma. METHODS: Using big-data analytics and artificial intelligence through the SAVANA Manager clinical platform, we analysed clinical data from patients with asthma from January 1 to May 10, 2020. RESULTS: Out of 71 182 patients with asthma, 1006 (1.41%) suffered from COVID-19. Compared to asthmatic individuals without COVID-19, patients with asthma and COVID-19 were significantly older (55 versus 42 years), predominantly female (66% versus 59%), smoked more frequently and had higher prevalence of hypertension, dyslipidaemias, diabetes and obesity. Allergy-related factors such as rhinitis and eczema were less common in asthmatic patients with COVID-19 (p<0.001). In addition, higher prevalence of these comorbidities was observed in patients with COVID-19 who required hospital admission. The use of inhaled corticosteroids (ICS) was lower in patients who required hospitalisation due to COVID-19, as compared to non-hospitalised patients (48.3% versus 61.5%; OR 0.58, 95% CI 0.44-0.77). Although patients treated with biologics (n=865; 1.21%) showed increased severity and more comorbidities at the ear, nose and throat level, COVID-19-related hospitalisations in these patients were relatively low (0.23%). CONCLUSION: Patients with asthma and COVID-19 were older and at increased risk due to comorbidity-related factors. ICS and biologics are generally safe and may be associated with a protective effect against severe COVID-19 infection.

Active Sleep Promotes Coherent Oscillatory Activity in the Cortico-Hippocampal System of Infant Rats.

Active sleep (AS) provides a unique developmental context for synchronizing neural activity within and between cortical and subcortical structures. In week-old rats, sensory feedback from myoclonic twitches, the phasic motor activity that characterizes AS, promotes coherent theta oscillations (4-8 Hz) in the hippocampus and red nucleus, a midbrain motor structure. Sensory feedback from twitches also triggers rhythmic activity in sensorimotor cortex in the form of spindle bursts, which are brief oscillatory events composed of rhythmic components in the theta, alpha/beta (8-20 Hz), and beta2 (20-30 Hz) bands. Here we ask whether one or more of these spindle-burst components are communicated from sensorimotor cortex to hippocampus. By recording simultaneously from whisker barrel cortex and dorsal hippocampus in 8-day-old rats, we show that AS, but not other behavioral states, promotes cortico-hippocampal coherence specifically in the beta2 band. By cutting the infraorbital nerve to prevent the conveyance of sensory feedback from whisker twitches, cortical-hippocampal beta2 coherence during AS was substantially reduced. These results demonstrate the necessity of sensory input, particularly during AS, for coordinating rhythmic activity between these two developing forebrain structures.

Proposed rituximab biosimilar BCD-020 versus reference rituximab for treatment of patients with indolent non-Hodgkin lymphomas: An international multicenter randomized trial.

BCD-020 is a proposed rituximab biosimilar, which has shown high similarity to rituximab in quality and nonclinical studies in vitro and in vivo. International multicenter clinical trial was conducted to compare efficacy and safety of BCD-020 and reference rituximab in adult (older than 18 years) patients with indolent lymphomas (follicular lymphoma grade 1-2, splenic marginal zone lymphoma, and nodal marginal zone lymphoma). Pharmacokinetics, pharmacodynamics, and immunogenicity were also studied. Patients with no previous biologic treatment for lymphoma were randomly assigned 1:1 to receive BCD-020 or comparator 375 mg/m2 for 4 weeks. Primary study outcome was day 50 overall response rate defined as complete or partial remission. Equivalence range was -20% to 20% for 95% CI for overall response rates difference. Secondary outcomes included adverse events, pharmacokinetics, pharmacodynamics, and immunogenicity. One hundred seventy-four patients were enrolled, 89 in BCD-020 arm and 85 in comparator arm. The overall response rate was 44.71% in BCD-020 arm and 41.89% in comparator arm. Limits of 95% confidence interval (CI) for difference of overall response rates between arms were (-12.62%-18.24%) showing equivalent efficacy. Sixty-one (68.54%) and 59 (69.41%) patients had at least one adverse event in BCD-020 arm or comparator arm, respectively. No unexpected adverse reactions were reported. Antidrug antibodies with no neutralizing activity were detected in two patients in comparator arm on day 14 further declining below detection threshold. Rituximab concentrations had equivalent pattern after intravenous administration of both drugs. Both drugs caused depletion of B-cells without significant influence on other blood cell lineages. In this study, we showed equivalent efficacy of BCD-020 and reference rituximab when used in patients with CD20-positive indolent lymphomas. We also confirmed pharmacokinetic equivalence of BCD-020 and reference rituximab. Safety profile, pharmacodynamics, and immunogenicity of BCD-020 were also comparable with those of reference rituximab.

Active Sleep Promotes Functional Connectivity in Developing Sensorimotor Networks.

A ubiquitous feature of active (REM) sleep in mammals and birds is its relative abundance in early development. In rat pups across the first two postnatal weeks, active sleep promotes the expression of synchronized oscillatory activity within and between cortical and subcortical sensorimotor structures. Sensory feedback from self-generated myoclonic twitches - which are produced exclusively during active sleep - also triggers neural oscillations in those structures. We have proposed that one of the functions of active sleep in early infancy is to provide a context for synchronizing developing structures. Specifically, neural oscillations contribute to a variety of neurodevelopmental processes, including synapse formation, neuronal differentiation and migration, apoptosis, and the refinement of topographic maps. In addition, synchronized oscillations promote functional connectivity between distant brain areas. Consequently, any condition or manipulation that restricts active sleep can, in turn, deprive the infant animal of substantial sensory experience, resulting in atypical developmental trajectories.

Iron bioavailability of four iron sources used to fortify infant cereals, using anemic weaning pigs as a model.

PURPOSE: Iron (Fe) deficiency anemia in young children is a global health concern which can be reduced by Fe fortification of foods. Cereal is often one of the first foods given to infants, providing adequate quantities of Fe during weaning. In this work, we have compared iron bioavailability and iron status of four iron sources used to fortify infant cereals, employing piglets as an animal model. METHOD: The study was conducted on 36 piglets, 30 of them with induced anemia. From day 28 of life, the weaned piglets were fed with four experimental diets (n = 6) each fortified with 120 mg Fe/kg by ferrous sulfate heptahydrate (FSH), electrolytic iron (EI), ferrous fumarate (FF), or micronized dispersible ferric pyrophosphate (MDFP) for another 21 days. In addition, one group of six anemic piglets fed with the basal diet with no iron added (Control-) and a Control+ group of non-anemic piglets (n = 6) were also studied. Blood indicators of iron status were measured after depletion and during the repletion period. The Fe content in organs, hemoglobin regeneration efficiency, and relative bioavailability (RBV) was also determined. RESULTS: The Fe salts adequately treated anemia in the piglets, allowing the animals to recover from the anemic state, although EI was less efficient with regard to replenishing Fe stores giving lower concentrations of plasma ferritin and iron in the spleen, liver, lung, and kidney. In addition, the RBV of EI was 88.27% with respect to the reference iron salt (FSH). CONCLUSIONS: Ferrous fumarate and MDFP were equally as bioavailable as the reference salt, and were used significantly better than EI in piglets. These results contribute to extend the evidence-based results for recommending the most suitable fortificant for infant cereals.

Stent optical inspection system calibration and performance.

Implantable medical devices, such as stents, have to be inspected 100% so no defective ones are implanted into a human body. In this paper, a novel optical stent inspection system is presented. By the combination of a high numerical aperture microscope, a triple illumination system, a rotational stage, and a CMOS camera, unrolled sections of the outer and inner surfaces of the stent are obtained with high resolution at high speed with a line-scan approach. In this paper, a comparison between the conventional microscope image formation and this new approach is shown. A calibration process and the investigation of the error sources that lead to inaccuracies of the critical dimension measurements are presented.

Cultivation of Solanum tuberosum in a former mining district for a safe human consumption integrating simulated digestion.

BACKGROUND: Potato (Solanum tuberosum) is a global crop and by far the most important non-cereal crop in the world. Therefore, it is necessary to assure its safe consumption. This is especially relevant in the case of its cultivation in abandoned mining areas, where the population tends to return to agriculture. In the present work, the objective is to evaluate the contribution to the diet of nutrients and contaminants of potato grown in soils from the Almadén area (mining district) by studying the intestinal absorption (in vitro) of the tuber, taking into account the preparation methods for its consumption. RESULTS: The results of contaminant and nutrient contents show that the potato peel retains significantly more elements (mainly in the case of toxic elements) than the flesh. Furthermore, potato (peel and flesh) is a good source of iron. CONCLUSION: It is recommended to boil potatoes with the peel in order to minimize nutrient loss and, before consumption, peeling them to eliminate possible risks due to contaminants. In addition, to minimize the risk due of mercury and to improve the levels of calcium, magnesium, potassium and sodium, it is recommended to add salt during the boiling process. © 2017 Society of Chemical Industry.

Sensorimotor processing in the newborn rat red nucleus during active sleep.

Sensory feedback from sleep-related myoclonic twitches is thought to drive activity-dependent development in spinal cord and brain. However, little is known about the neural pathways involved in the generation of twitches early in development. The red nucleus (RN), source of the rubrospinal tract, has been implicated in the production of phasic motor activity during active sleep in adults. Here we hypothesized that the RN is also a major source of motor output for twitching in early infancy, a period when twitching is an especially abundant motor behavior. We recorded extracellular neural activity in the RN during sleep and wakefulness in 1-week-old unanesthetized rats. Neurons in the RN fired phasically before twitching and wake movements of the contralateral forelimb. A subpopulation of neurons in the RN exhibited a significant peak of activity after forelimb movement onset, suggesting reafferent sensory processing. Consistent with this observation, manual stimulation of the forelimb evoked RN responses. Unilateral inactivation of the RN using a mixture comprising GABAA, GABAB, and glycine receptor agonists caused an immediate and temporary increase in motor activity followed by a marked and prolonged decrease in twitching and wake movements. Altogether, these data support a causal role for the RN in infant motor behavior. Furthermore, they indicate that twitching, which is characterized by discrete motor output and reafferent input, provides an opportunity for sensorimotor integration and activity-dependent development of topography within the newborn RN.

Spontaneous activity and functional connectivity in the developing cerebellorubral system.

The development of the cerebellar system depends in part on the emergence of functional connectivity in its input and output pathways. Characterization of spontaneous activity within these pathways provides insight into their functional status in early development. In the present study we recorded extracellular activity from the interpositus nucleus (IP) and its primary downstream target, the red nucleus (RN), in unanesthetized rats at postnatal days 8 (P8) and P12, a period of dramatic change in cerebellar circuitry. The two structures exhibited state-dependent activity patterns and age-related changes in rhythmicity and overall firing rate. Importantly, sensory feedback (i.e., reafference) from myoclonic twitches (spontaneous, self-generated movements that are produced exclusively during active sleep) drove neural activity in the IP and RN at both ages. Additionally, anatomic tracing confirmed the presence of cerebellorubral connections as early as P8. Finally, inactivation of the IP and adjacent nuclei using the GABAA receptor agonist muscimol caused a substantial decrease in neural activity in the contralateral RN at both ages, as well as the disappearance of rhythmicity; twitch-related activity in the RN, however, was preserved after IP inactivation, indicating that twitch-related reafference activates the two structures in parallel. Overall, the present findings point to the contributions of sleep-related spontaneous activity to the development of cerebellar networks.

A valuable and promising method for recording brain activity in behaving newborn rodents.

Neurophysiological recording of brain activity has been critically important to the field of neuroscience, but has contributed little to the field of developmental psychobiology. The reasons for this can be traced largely to methodological difficulties associated with recording neural activity in behaving newborn rats and mice. Over the last decade, however, the evolution of methods for recording from head-fixed newborns has heralded a new era in developmental neurophysiology. Here, we review these recent developments and provide a step-by-step primer for those interested in applying the head-fix method to their own research questions. Until now, this method has been used primarily to investigate spontaneous brain activity across sleep and wakefulness, the contributions of the sensory periphery to brain activity, or intrinsic network activity. Now, with some ingenuity, the uses of the head-fix method can be expanded to other domains to benefit our understanding of brain-behavior relations under normal and pathophysiological conditions across early development.

Interventional nephrology in Puerto Rico: a four year experience.

Puerto Rico is one of the most prevalent areas covered by Medicare in need of renal replacement therapy for which interventional procedures are performed. A cumulative analysis of this management is reported in patients during the period between June 2007 and August 2010. Experience accumulated with 3755 surgical patients revealed that 58% had intravascular catheters, 28% had arteriovenous fistulas, 15% had arteriovenous grafts, and 2% without vascular access. Procedures performed in these patients were: catheter introduction in 1990 cases (33%), angioplasty in 751 cases (20%), angiography in 450 cases (12%), thrombectomy in 413 cases (11%) and venous mapping in 151 cases (4%). The success rates of these procedures were evaluated by analysis of the Society of Interventional Radiology (SIR) criteria for Lifeline Vascular Access. Using SIR definition of success rate for at least one session that includes "declots", placement of catheters and angioplasty, our results revealed an average of 98.2% overall success rate greater than the standard value KDOQI/SIR (> 85% ). This study has documented for four years the success rate of Vascular Interventional Nephrology Center at Auxilio Mutuo Hospital. In order to maintain this success rate is necessary to further evaluate its effectiveness and, most importantly, the development of an educational program for vascular access in patients with chronic kidney disease prior to placement in dialysis units.

Effect of pine bark extract and its phenolic compounds on selected pathogenic and probiotic bacterial strains.

Introduction: Inflammatory bowel disease (IBD) comprises a heterogeneous group of chronic diseases as ulcerative colitis (UC) and Crohn's disease (CD). IBD is the result of a dysregulation of intestinal homeostasis with a host's loss of tolerance toward normal enteric microflora. Plant-based extracts as phenolic compounds can play a role by modulating the intestinal inflammation response. Methods: The in vitro antimicrobial activity of French maritime pine bark extract (PBE) and its phenolic constituents has been investigated in this study. Furthermore, the ability of PBE and phenolic compounds (caffeic acid, chlorogenic acid, ferulic acid, gallic acid and taxifolin) to modulate the microbiota has been assessed. Results: Phenolic compounds and PBE showed a great inhibitory effect on the pathogens growth at the highest concentration assessed (1.25 mg/mL). The growth of E. sakazakii and E. faecalis were affected by the effect of caffeic acid and ferulic acid. Taxifolin showed a very strong activity against Listeria sp. (with a reduction ~98%). Gallic acid revealed antibacterial effect on S. aureus at different concentrations. The inhibitory effect of PBE was highly significant on the growth of E. coli O157:H7. PBE, caffeic acid and chlorogenic acid seem to provide the greatest beneficial effect on the probiotic bacteria. However, the highest concentrations of taxifolin may have impaired the growth of beneficial microbiota. Conclusion: Present findings could be of interest for considering PBE and/or its phenolic constituents as protectors against gastrointestinal disturbances which lead to ulcerative colitis and Crohn's disease.

Development of LuxIA, a Cloud-Based AI Diabetic Retinopathy Screening Tool Using a Single Color Fundus Image.

Purpose: Diabetic retinopathy (DR) is the leading cause of vision impairment in working-age adults. Automated screening can increase DR detection at early stages at relatively low costs. We developed and evaluated a cloud-based screening tool that uses artificial intelligence (AI), the LuxIA algorithm, to detect DR from a single fundus image. Methods: Color fundus images that were previously graded by expert readers were collected from the Canarian Health Service (Retisalud) and used to train LuxIA, a deep-learning-based algorithm for the detection of more than mild DR. The algorithm was deployed in the Discovery cloud platform to evaluate each test set. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were computed using a bootstrapping method to evaluate the algorithm performance and compared through different publicly available datasets. A usability test was performed to assess the integration into a clinical tool. Results: Three separate datasets, Messidor-2, APTOS, and a holdout set from Retisalud were evaluated. Mean sensitivity and specificity with 95% confidence intervals (CIs) reached for these three datasets were 0.901 (0.901-0.902) and 0.955 (0.955-0.956), 0.995 (0.995-0.995) and 0.821 (0.821-0.823), and 0.911 (0.907-0.912) and 0.880 (0.879-0.880), respectively. The usability test confirmed the successful integration of LuxIA into Discovery. Conclusions: Clinical data were used to train the deep-learning-based algorithm LuxIA to an expert-level performance. The whole process (image uploading and analysis) was integrated into the cloud-based platform Discovery, allowing more patients to have access to expert-level screening tools. Translational Relevance: Using the cloud-based LuxIA tool as part of a screening program may give diabetic patients greater access to specialist-level decisions, without the need for consultation.

Symptoms timeline and outcomes in amyotrophic lateral sclerosis using artificial intelligence.

Amyotrophic lateral sclerosis (ALS) is a fatal, neurodegenerative motor neuron disease. Although an early diagnosis is crucial to provide adequate care and improve survival, patients with ALS experience a significant diagnostic delay. This study aimed to use real-world data to describe the clinical profile and timing between symptom onset, diagnosis, and relevant outcomes in ALS. Retrospective and multicenter study in 5 representative hospitals and Primary Care services in the SESCAM Healthcare Network (Castilla-La Mancha, Spain). Using Natural Language Processing (NLP), the clinical information in electronic health records of all patients with ALS was extracted between January 2014 and December 2018. From a source population of all individuals attended in the participating hospitals, 250 ALS patients were identified (61.6% male, mean age 64.7 years). Of these, 64% had spinal and 36% bulbar ALS. For most defining symptoms, including dyspnea, dysarthria, dysphagia and fasciculations, the overall diagnostic delay from symptom onset was 11 (6-18) months. Prior to diagnosis, only 38.8% of patients had visited the neurologist. In a median post-diagnosis follow-up of 25 months, 52% underwent gastrostomy, 64% non-invasive ventilation, 16.4% tracheostomy, and 87.6% riluzole treatment; these were more commonly reported (all Ps < 0.05) and showed greater probability of occurrence (all Ps < 0.03) in bulbar ALS. Our results highlight the diagnostic delay in ALS and revealed differences in the clinical characteristics and occurrence of major disease-specific events across ALS subtypes. NLP holds great promise for its application in the wider context of rare neurological diseases.

Development of a predictive model of venous thromboembolism recurrence in anticoagulated cancer patients using machine learning.

INTRODUCTION: Patients with cancer and venous thromboembolism (VTE) show a high risk of VTE recurrence during anticoagulant treatment. This study aimed to develop a predictive model to assess the risk of VTE recurrence within 6 months at the moment of primary VTE diagnosis in these patients. MATERIALS AND METHODS: Using the EHRead® technology, based on Natural Language Processing (NLP) and machine learning (ML), the unstructured data in electronic health records from 9 Spanish hospitals between 2014 and 2018 were extracted. Both clinically- and ML-driven feature selection were performed to identify predictors for VTE recurrence. Logistic regression (LR), decision tree (DT), and random forest (RF) algorithms were used to train different prediction models, which were subsequently validated in a hold-out data set. RESULTS: A total of 16,407 anticoagulated cancer patients with diagnosis of VTE were identified (54.4 % male and median age 70). Deep vein thrombosis, pulmonary embolism and metastases were observed in 67.2 %, 26.6 %, and 47.7 % of the patients, respectively. During the study follow-up, 11.4 % of the patients developed a recurrent VTE, being more frequent in patients with lung cancer. Feature selection and model training based on ML identified primary pulmonary embolism, deep vein thrombosis, metastasis, adenocarcinoma, hemoglobin and serum creatinine levels, platelet and leukocyte count, family history of VTE, and patients' age as predictors of VTE recurrence within 6 months of VTE diagnosis. The LR model had an AUC-ROC (95 % CI) of 0.66 (0.61, 0.70), the DT of 0.69 (0.65, 0.72) and the RF of 0.68 (0.63, 0.72). CONCLUSIONS: This is the first ML-based predictive model designed to predict 6-months VTE recurrence in patients with cancer. These results hold great potential to assist clinicians to identify the high-risk patients and improve their clinical management.

How can artificial intelligence optimize value-based contracting?

Efforts in the pharmaceutical market have been aimed at ensuring that the benefits obtained from the introduction of new therapies justify the associated costs. In recent years, drug payment models in healthcare have undergone a dramatic shift from focusing on volume (i.e., size of the target clinical population) to focusing on value (i.e., drug performance in real-world settings). In this context, value-based contracts (VBCs) were designed to align the payment of a drug to its clinical performance outside clinical trials by evaluating the effectiveness using real-word evidence (RWE). Despite their widespread implementation, different factors jeopardize the application of VBCs to most marketed drugs in a near future, including the need for easily measurable and relevant outcomes associated with clinical improvements, and access to a large patient population to assess said outcomes. Here, we argue that the extraction and analysis of massive amounts of RWE captured in patients' electronic health records (EHRs) will circumvent these issues and optimize negotiations in VBCs. Particularly, the use of Natural Language Processing (NLP) has proven successful in the analysis of structured and unstructured clinical information in EHRs in multicenter research studies. Thus, the application of NLP to analyze patient-centered information in EHRs in the context of innovative contracting can be utterly beneficial as it enables the real-time evaluation of treatment response and financial impact in real-world settings.