RESUMEN

In radiotherapy treatment planning, optimization is essential for achieving the most favorable plan by adjusting optimization criteria. This study introduced an innovative approach to automatically fine-tune optimization parameters for volumetric modulated arc therapy prostate planning, ensuring all constraints were met. A knowledge-based planning model was invoked, and the fine-tuning process was applied through an in-house developed script. Among 25 prostate plans, this fine-tuning increased the number of plans meeting all constraints from 10/25 to 22/25, with a reduction in mean monitor units per gray without increasing plan's complexity. This automation improved efficiency by saving time and resources in treatment planning.

RESUMEN

Background: Integrating health information into university information systems holds significant potential for enhancing student support and well-being. Despite the growing body of research highlighting issues faced by university students, including stress, depression, and disability, little has been done in the informatics field to incorporate health technologies at the institutional level. Objective: This study aims to investigate the current state of health information integration within university systems and provide design recommendations to address existing gaps and opportunities. Methods: We used a user-centered approach to conduct interviews and focus group sessions with stakeholders to gather comprehensive insights and requirements for the system. The methodology involved data collection, analysis, and the development of a suggested workflow. Results: The findings of this study revealed the shortcomings in the current process of handling health and disability data within university information systems. In our results, we discuss some requirements identified for integrating health-related information into student information systems, such as privacy and confidentiality, timely communication, task automation, and disability resources. We propose a workflow that separates the process into 2 distinct components: a health and disability system and measures of quality of life and wellness. The proposed workflow highlights the vital role of academic advisors in facilitating support and enhancing coordination among stakeholders. Conclusions: To streamline the workflow, it is vital to have effective coordination among stakeholders and redesign the university information system. However, implementing the new system will require significant capital and resources. We strongly emphasize the importance of increased standardization and regulation to support the information system requirements for health and disability. Through the adoption of standardized practices and regulations, we can ensure the smooth and effective implementation of the required support system.

Asunto(s)

RESUMEN

INTRODUCTION: An on-road study was conducted to examine the effects of level 2 automation on the stressfulness and enjoyment of driving and driving attention following prolonged usage. The study also examined the changes in the automated driving experience and attention over time as well as important predictors such as pre-driving trust in technology and attitudes toward automated systems. METHOD: Motorists who had never used automated systems drove a level 2 automation vehicle for a 6-8 week period. RESULTS: Participants reported that the automated systems reduced the stress of driving and made traveling more enjoyable and relaxing. They also reported that the automation did not make traveling boring and take the fun out of driving. Participants indicated that their minds tended to wander when the automation was operating. The stressfulness of the automated driving experience decreased over time. Participants also reported feeling increasingly comfortable driving with the automation without monitoring it closely. The enjoyment and stress of automated driving is important because it shapes the willingness to use the automation and, hence, the safeness of driving. As expected, intentions to use and purchase automated systems were strongly predicted by the perceived favorableness of driving with the automation. Participants' pre-driving beliefs about automated systems, rather than their trust, appears to have shaped their experiences with the automation. PRACTICAL APPLICATIONS: Although some of the findings suggest that automated systems increase unsafe behavior by novice users, other facets of the surveys suggest that motorists are cognizant of the risks of automated driving and discreet in their usage of the automation.

Asunto(s)

RESUMEN

In this study, we adapted an HP D100 Single Cell Dispenser - a novel low-cost thermal inkjet (TIJ) platform with impedance-based single cell detection - for dispensing of individual cells and one-pot sample preparation. We repeatedly achieved label-free identification of up to 1,300 proteins from a single cell in a single run using an Orbitrap Fusion Lumos Mass Spectrometer coupled to either an Acquity UPLC M-class system or a Vanquish Neo UHPLC system. The developed sample processing workflow is highly reproducible, robust, and applicable to standardized 384- and 1536-well microplates, as well as glass LC vials. We demonstrate the applicability of the method for proteomics of single cells from multiple cell lines, mixed cell suspensions, and glioblastoma tumor spheroids. As additional proof of robustness, we monitored the results of genetic manipulations and the expression of engineered proteins in individual cells. Our cost-effective and robust single-cell proteomics workflow can be transferred to other labs interested in studying cells at the individual cell level.

RESUMEN

The pharmaceutical industry is increasingly embracing laboratory automation to enhance experimental efficiency and operational resilience, particularly through the integration of automated liquid handlers (ALHs). This paper explores the integration of the low-cost Opentrons OT-2 liquid handling robot with F. Hoffmann-La Roche AG's in-house workflow orchestration software, AutoLab, to overcome barriers to lab automation. By leveraging the OT-2's development-oriented interfaces and AutoLab's modular architecture, we achieved a user-friendly, cost-efficient, and flexible automation solution that aligns with FAIR (findable, accessible, interoperable, reusable) data principles. We demonstrate an advanced workflow development methodology, utilizing the software architecture, that facilitates the creation of two flexible pipetting protocols and medium complexity assays. This deep integration approach diminishes the learning curve for novice users while simultaneously enhancing the overall efficiency and reliability of the experimental workflow. Our findings suggest that such integrations can significantly mitigate the challenges associated with lab automation, including cost, complexity, and adaptability, paving the way for more accessible and robust automated systems in pharmaceutical research.

RESUMEN

With the growing role of artificial intelligence (AI) in our lives, attention is increasingly turning to the way that humans and AI work together. A key aspect of human-AI collaboration is how people integrate judgements or recommendations from machine agents, when they differ from their own judgements. We investigated trust in human-machine teaming using a perceptual judgement task based on the judge-advisor system. Participants ( n = 89 ) estimated a perceptual quantity, then received a recommendation from a machine agent. The participants then made a second response which combined their first estimate and the machine's recommendation. The degree to which participants shifted their second response in the direction of the recommendations provided a measure of their trust in the machine agent. We analysed the role of advice distance in people's willingness to change their judgements. When a recommendation falls a long way from their initial judgement, do people come to doubt their own judgement, trusting the recommendation more, or do they doubt the machine agent, trusting the recommendation less? We found that although some participants exhibited these behaviours, the most common response was neither of these tendencies, and a simple model based on averaging accounted best for participants' trust behaviour. We discuss implications for theories of trust, and human-machine teaming.

Asunto(s)

RESUMEN

Computer Aided Detection (CAD) has been used to help readers find cancers in mammograms. Although these automated systems have been shown to help cancer detection when accurate, the presence of CAD also leads to an over-reliance effect where miss errors and false alarms increase when the CAD system fails. Previous research investigated CAD systems which overlayed salient exogenous cues onto the image to highlight suspicious areas. These salient cues capture attention which may exacerbate the over-reliance effect. Furthermore, overlaying CAD cues directly on the mammogram occludes sections of breast tissue which may disrupt global statistics useful for cancer detection. In this study we investigated whether an over-reliance effect occurred with a binary CAD system, which instead of overlaying a CAD cue onto the mammogram, reported a message alongside the mammogram indicating the possible presence of a cancer. We manipulated the certainty of the message and whether it was presented only to indicate the presence of a cancer, or whether a message was displayed on every mammogram to state whether a cancer was present or absent. The results showed that although an over-reliance effect still occurred with binary CAD systems miss errors were reduced when the CAD message was more definitive and only presented to alert readers of a possible cancer.

Asunto(s)

RESUMEN

OBJECTIVE: We investigate the impact of event uncertainty, decision support (DS) display format, and DS sensitivity on participants' behavior, performance, subjective workload, and perception of DS usefulness and performance in a classification task. BACKGROUND: DS systems can positively and negatively affect decision accuracy, performance time, and workload. The ability to access DS selectively, based on informational needs, might improve DS effectiveness. METHOD: Participants performed a sensory classification task in which they were allowed to request DS on a trial-by-trial basis. DS was presented in separated-binary (SB), separated-likelihood (SL), or integrated-likelihood (IL) formats. Access preferences, task performance, performance time, subjective workload, and perceived DS usefulness and performance were recorded. RESULTS: Participants accessed DS more often when it was highly sensitive, stimulus information was highly uncertain, or the DS cue and stimulus information were perceptually integrated. Effective sensitivity was highest with the integrated likelihood DS. Although the separated likelihood DS provided more information than the binary likelihood DS, it was accessed less often, leading to lower sensitivity. CONCLUSION: Participants are most likely to access DS when raw stimulus information is highly uncertain and appear to make effective use of likelihood DS only when DS cues are integrated with raw stimulus information within a display. APPLICATION: Results suggest that DS use will be most effective when likelihood DS cues and raw stimulus information are perceptually integrated. When DS cues and raw stimuli cannot be perceptually integrated, binary cues from the DS will be more effective than likelihood cues.

RESUMEN

Fruits are mature ovaries of flowering plants that are integral to human diets, providing essential nutrients such as vitamins, minerals, fiber and antioxidants that are crucial for health and disease prevention. Accurate classification and segmentation of fruits are crucial in the agricultural sector for enhancing the efficiency of sorting and quality control processes, which significantly benefit automated systems by reducing labor costs and improving product consistency. This paper introduces the "FruitSeg30_Segmentation Dataset & Mask Annotations", a novel dataset designed to advance the capability of deep learning models in fruit segmentation and classification. Comprising 1969 high-quality images across 30 distinct fruit classes, this dataset provides diverse visuals essential for a robust model. Utilizing a U-Net architecture, the model trained on this dataset achieved training accuracy of 94.72 %, validation accuracy of 92.57 %, precision of 94 %, recall of 91 %, f1-score of 92.5 %, IoU score of 86 %, and maximum dice score of 0.9472, demonstrating superior performance in segmentation tasks. The FruitSeg30 dataset fills a critical gap and sets new standards in dataset quality and diversity, enhancing agricultural technology and food industry applications.

RESUMEN

Zebrafish is a highly advantageous model animal for drug screening and toxicity evaluation thanks to its amenability to optical imaging (i.e., transparency), possession of organ structures similar to humans, and the ease with which disease models can be established. However, current zebrafish drug screening technologies and devices suffer from limitations such as low level of automation and throughput, and low accuracy caused by the heterogeneity among individual zebrafish specimens. To address these issues, we herein develop a high-throughput zebrafish drug screening system. This system is capable of maintaining optimal culturing conditions and simultaneously monitoring and analyzing the movement of 288 zebrafish larvae under various external conditions, such as drug combinations. Moreover, to eliminate the effect of heterogeneity, locomotion of participating zebrafish is assessed and grouped before experiments. It is demonstrated that in contrast to the experimental results without pre-selection, which shows ∼20 % damaged motor function (i.e., degree of attenuation), the drug-induced variations among zebrafish with equivalent mobility reaches ∼80 %. Overall, our high-throughput zebrafish drug screening system overcomes current limitations by improving automation, throughput, and accuracy, resulting in enhanced detection of drug-induced variations in zebrafish motor function.

RESUMEN

Advancements in radiotherapy auto-segmentation necessitate reliable and efficient workflows. Therefore, a standardized fully automatic workflow was developed for three commercially available deep learning-based auto-segmentation applications and compared to a manual workflow for safety and efficiency. The workflow underwent safety evaluation with failure mode and effects analysis. Notably, eight failure modes were reduced, including seven with severity factors ≥7, indicating the effect on patients, and two with Risk Priority Number value >125, which assesses relative risk level. Efficiency, measured by mouse clicks, showed zero clicks with the automatic workflow. This automation illustrated improvement in both safety and efficiency of workflow.

RESUMEN

The non-linear complex relationships among the process variables in wastewater and waste gas treatment systems possess a significant challenge for real-time systems modelling. Data driven artificial intelligence (AI) tools are increasingly being adopted to predict the process performance, cost-effective process monitoring, and the control of different waste treatment systems, including those involving resource recovery. This review presents an in-depth analysis of the applications of emerging AI tools in physico-chemical and biological processes for the treatment of air pollutants, water and wastewater, and resource recovery processes. Additionally, the successful implementation of AI-controlled wastewater and waste gas treatment systems, along with real-time monitoring at the industrial scale are discussed.

RESUMEN

Background: In the United States, patients with monochorionic diamniotic twins who undergo in utero fetoscopic laser photocoagulation (FLP) for twin-twin transfusion syndrome (TTTS) may travel great distances for care. After delivery, many parents cannot return to study sites for formal pediatric evaluation due to geographic location and cost. Objective: The aim of this study was to collect long-term pediatric outcomes in patients who underwent FLP for TTTS. Methods: We assessed the feasibility of using a web-based survey designed in REDCap (Research Electronic Data Capture; Vanderbilt University) to collect parent-reported outcomes in children treated for TTTS at a single center during 2011-2019. Patients with ≥1 neonatal survivor were invited via email to complete 5 possible questionnaires: the child status questionnaire (CSQ); fetal center questionnaire (FCQ); Ages & Stages Questionnaires, Third Edition (ASQ-3); Modified Checklist for Autism in Toddlers, Revised With Follow-Up (M-CHAT-R/F); and thank you questionnaire (TYQ). The R programming language (R Foundation for Statistical Computing) was used to automate survey distribution, scoring, and creation of customized reports. The survey was performed in 2019 and repeated after 12 months in the same study population in 2020. Results: A total of 389 patients in 26 different states and 2 international locations had an email address on file and received an invitation in 2019 to complete the survey (median pediatric age 48.9, IQR 1.0-93.6 months). Among surveyed mothers in 2019, the overall response rate was 37.3% (145/389), and the questionnaire completion rate was 98% (145/148), 87.8% (130/148), 71.1% (81/100), 86.4% (19/22), and 74.3% (110/148) for the CSQ, FCQ, ASQ-3, M-CHAT-R/F, and TYQ, respectively. In 2020, the overall response rate was 57.8% (56/97), and the questionnaire completion rate was 96.4% (54/56), 91.1% (51/56), 86.1% (31/36), 91.7% (11/12), and 80.4% (45/56) for the CSQ, FCQ, ASQ-3, M-CHAT-R/F, and TYQ, respectively. Conclusions: This is the first study to use both REDCap and computer automation to aid in the dissemination, collection, and reporting of surveys to collect long-term pediatric outcomes in the field of fetal medicine.

RESUMEN

Recent developments in affordable depth imaging hardware and the use of 2D Convolutional Neural Networks (CNN) in object detection and segmentation have accelerated the adoption of machine vision in a range of applications, with mainstream models often out-performing previous application-specific architectures. The need for the release of training and test datasets with any work reporting model development is emphasized to enable the re-evaluation of published work. An additional reporting need is the documentation of the performance of the re-training of a given model, quantifying the impact of stochastic processes in training. Three mango orchard applications were considered: the (i) fruit count, (ii) fruit size and (iii) branch avoidance in automated harvesting. All training and test datasets used in this work are available publicly. The mAP 'coefficient of variation' (Standard Deviation, SD, divided by mean of predictions using models of repeated trainings × 100) was approximately 0.2% for the fruit detection model and 1 and 2% for the fruit and branch segmentation models, respectively. A YOLOv8m model achieved a mAP50 of 99.3%, outperforming the previous benchmark, the purpose-designed 'MangoYOLO', for the application of the real-time detection of mango fruit on images of tree canopies using an edge computing device as a viable use case. YOLOv8 and v9 models outperformed the benchmark MaskR-CNN model in terms of their accuracy and inference time, achieving up to a 98.8% mAP50 on fruit predictions and 66.2% on branches in a leafy canopy. For fruit sizing, the accuracy of YOLOv8m-seg was like that achieved using Mask R-CNN, but the inference time was much shorter, again an enabler for the field adoption of this technology. A branch avoidance algorithm was proposed, where the implementation of this algorithm in real-time on an edge computing device was enabled by the short inference time of a YOLOv8-seg model for branches and fruit. This capability contributes to the development of automated fruit harvesting.

Asunto(s)

RESUMEN

Direct fluorination of a tosylate or mesylate precursor has been a wide-spread and reliable way for radio-fluorination. This approach can be difficult to achieve when the precursor cannot be easily obtained or is chemically unstable. A possible alternative method is to radiolabel ethylene ditosylate or 1,3-propanediol di-p-tosylate to form a radiofluorinated synthon. Here we present the automation of a simplified and reliable approach for the two-step fluorination using [18F]FP-TMP, an analog of antibacterial agent trimethoprim. We demonstrate the feasibility of purifying the fluorinated synthon via filtration, and one final HPLC purification on a commercially available Trasis AllinOne module. The overall reaction time for the two-step reaction is around 90 min andthe decay-corrected yield for more than fifty preparations of [18F]FP-TMP is 22 ± 5 % with high radiochemical purity (≥ 90 %) and specific activities (147 ± 107 GBq/µmol). All batches passed pre-established quality control specifications, demonstrating the utility of using this method in tracer syntheses that meet good manufacturing practice (GMP) requirement. This method can be adopted to the syntheses of other radiotracers, such as [18F]FE-TMP, (+)-[18F]F-PHNO and [18F]FFMZ.

RESUMEN

BACKGROUND AND PURPOSE: Over the past decade, tools for automation of various sub-tasks in radiotherapy planning have been introduced, such as auto-contouring and auto-planning. The purpose of this study was to benchmark what degree of automation is possible. MATERIALS AND METHODS: A challenge to perform automated treatment planning for prostate and prostate bed radiotherapy was set up. Participants were provided with simulation CTs and a treatment prescription and were asked to use automated tools to produce a deliverable radiotherapy treatment plan with as little human intervention as possible. Plans were scored for their adherence to the protocol when assessed using consensus expert contours. RESULTS: Thirteen entries were received. The top submission adhered to 81.8% of the minimum objectives across all cases using the consensus contour, meeting all objectives in one of the ten cases. The same system met 89.5% of objectives when assessed with their own auto-contours, meeting all objectives in four of the ten cases. The majority of systems used in the challenge had regulatory clearance (Auto-contouring: 82.5%, Auto-planning: 77%). Despite the 'hard' rule that participants should not check or edit contours or plans, 69% reported looking at their results before submission. CONCLUSIONS: Automation of the full planning workflow from simulation CT to deliverable treatment plan is possible for prostate and prostate bed radiotherapy. While many generated plans were found to require none or minor adjustment to be regarded as clinically acceptable, the result indicated there is still a lack of trust in such systems preventing full automation.

RESUMEN

The use of organoid models in biomedical research has grown substantially since their inception. As they gain popularity among scientists seeking more complex and biologically relevant systems, there is a direct need to expand and clarify potential uses of such systems in diverse experimental contexts. Herein we outline a high-content screening (HCS) platform that allows researchers to screen drugs or other compounds against three-dimensional (3D) cell culture systems in a multi-well format (384-well). Furthermore, we compare the quality of robotic liquid handling with manual pipetting and characterize and contrast the phenotypic effects detected by confocal imaging and biochemical assays in response to drug treatment. We show that robotic liquid handling is more consistent and amendable to high throughput experimental designs when compared to manual pipetting due to improved precision and automated randomization capabilities. We also show that image-based techniques are more sensitive to detecting phenotypic changes within organoid cultures than traditional biochemical assays that evaluate cell viability, supporting their integration into organoid screening workflows. Finally, we highlight the enhanced capabilities of confocal imaging in this organoid screening platform as they relate to discerning organoid drug responses in single-well co-cultures of organoids derived from primary human biopsies and patient-derived xenograft (PDX) models. Altogether, this platform enables automated, imaging-based HCS of 3D cellular models in a non-destructive manner, opening the path to complementary analysis through integrated downstream methods.

RESUMEN

Accurate quantification of von Willebrand factor ristocetin cofactor activity (VWF:RCo) is critical for the diagnosis and classification of von Willebrand disease, the most common hereditary and acquired bleeding disorder in humans. Moreover, it is important to accurately assess the function of von Willebrand factor (VWF) concentrates within the pharmaceutical industry to provide consistent and high-quality biopharmaceuticals. Although the performance of VWF:RCo assay has been improved by using coagulation analyzers, which are specialized devices for blood and blood plasma samples, scientists still report a high degree of intra- and inter-assay variation in clinical laboratories. Moreover, high, manual sample dilutions are required for VWF:RCo determination of VWF concentrates within the pharmaceutical industry, which are a major source for assay imprecision. For the first time, we present a precise and accurate method to determine VWF:RCo, where all critical pipetting and mixing steps are automated. A pre-dilution setup was established on CyBio FeliX (Analytik-Jena) liquid handling system, and an adapted VWF:RCo method on BCS-XP analyzer (Siemens) is used. The automated pre-dilution method was executed on three different, most frequently used coagulation analyzers and compared to manual pre-dilutions performed by an experienced operator. Comparative sample testing revealed a similar assay precision (coefficient of variation = 5.9% automated, 3.1% manual pre-dilution) and no significant differences between the automated approach and manual dilutions of an expert in this method. While no outliers were generated with the automated procedure, the manual pre-dilution resulted in an error rate of 8.3%. Overall, this operator-independent protocol enables standardization and offers an efficient way of fully automating VWF activity assays, while maintaining the precision and accuracy of an expert analyst. Key features • Automated pre-dilution setup for von Willebrand factor concentrates of various natures. • Combination of a liquid handling system (CyBio FeliX) with a coagulation analyzer (BCS-XP). • Simplifies method transfer to other laboratories. • Basic training for CyBio FeliX and BCS-XP is required. Graphical overview VWF:RCo assay principle and measurement setup. Platelets (yellow ellipsoids) with negative surface charge (- - -) are treated with formaldehyde, which partly denatures the cell surface and thus stabilizes platelets for use as assay reagents. Stabilized platelets (dark-yellow-framed yellow ellipsoids) are then brought in contact with ristocetin A (chemical structure shown; black dots), which binds to the platelet surface and facilitates binding of VWF (green circles). The graphs show an example of quantitative determination of platelet agglutination by measurement of light transmission, where increasing amounts of VWF increase light transmission over time. The photo in the left-bottom corner shows the CyBio FeliX setup for VWF sample dilution and the photo in the right-bottom corner displays the BCS-XP system, which is used for VWF:RCo measurements.

RESUMEN

Driven by demographic changes and dwindling Science Technology Engineering Mathematics enrolments, our research introduces no-code automation as a strategic response, aimed at mitigating labor shortages while enhancing productivity and safety in the laboratory environment. Employing a user-friendly, no-code software platform, we automated a complex HPTLC assay, enabling laboratory personnel to configure and modify workflows without requiring specialized programming skills. The manuscript outlines the deployment of a collaborative robot (cobot), a programmable logic controller (PLC), and the utilization of self-developed open-source hardware components to establish automated stations for sample handling, incubation, spraying, detection, and storage within the assay process. The research addresses challenges such as the handling of fragile HPTLC plates and the seamless integration of automated stations, solved through innovative design solutions and adaptive programming methods. This investigation demonstrates the feasibility and efficiency of no-code automation in overcoming skilled labor deficits.