•This work use metrics (priority weighted fixed issues, versatility and breadth index metrics and developer average bug fixing time) to evaluate developer individual performance and contribution ...assessment for bug fixing.•These metrics are applied on three bug reports features to generate developer expertise score which are priority, product name and bug time-stamp.•The proposed approach demonstrates that developer expertise plays an important role to fix the bug within minimum time and to reduce bug tossing length.•Calculate DES accuracy of reassignment and hit ratio.•DES comparison is done with other state-of-arts as well as against machine learning based bug triaging approaches namely: Naïve bayes, support vector machine and C4.5.
Display omitted
Existing bug triage approaches for developer recommendation systems are mainly based on machine learning (ML) techniques. These approaches have shown low prediction accuracy and high bug tossing length (BTL).
The objective of this paper is to develop a robust algorithm for reducing BTL based on the concept of developer expertise score (DES).
None of the existing approaches to the best of our knowledge have utilized metrics to build developer expertise score. The novel strategy of DES is consisted of two stages: Stage-I consisted of an offline process for detecting the developers based on DES which computes the score using priority, versatility and average fix-time for his individual contributions. The online system process consisted of finding the capable developers using three kinds of similarity measures (feature-based, cosine-similarity and Jaccard). Stage-II of the online process consisted of simply ranking the developers. Hit-ratio and reassignment accuracy were used for performance evaluation. We compared our system against the ML-based bug triaging approaches using three types of classifiers: Navies Bayes, Support Vector Machine and C4.5 paradigms.
By adapting the five open source databases, namely: Mozilla, Eclipse, Netbeans, Firefox, and Freedesktop, covering 41,622 bug reports, our novel DES system yielded a mean accuracy, precision, recall rate and F-score of 89.49%, 89.53%, 89.42% and 89.49%, respectively, reduced BTLs of up to 88.55%. This demonstrates an improvement of up to 20% over existing strategies.
This work presented a novel developer recommendation algorithm to rank the developers based on a metric-based integrated score for bug triaging. This integrated score was based on the developer's expertise with an objective to improve (i) bug assignment and (ii) reduce the bug tossing length. Such architecture has an application in software bug triaging frameworks.
•A review on software bug triaging automation using AI is presented.•Major AI technologies are discussed for automation of software bug triaging.•Popular performance parameters for AI techniques for ...automation of SBT is discussed.•Management of vast bug information requires automation of software bug triaging.•Challenges and possible future works in utilizing AI for automating SBT are discussed.
The timely release of defect-free software and the optimization of development costs depend on efficient software bug triaging (SBT) techniques. SBT can also help in managing the vast information available in software bug repositories. Recently, Artificial Intelligence (AI)-based emerging technologies have been utilized excessively, however, it is not clear how it is shaping the design, development, and performance in the field of SBT. It is therefore important to write this well-planned, comprehensive, and timely needed AI-based SBT review, establishing clear findings. For selecting the key studies in SBT, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) analysis was carried out, and 123 studies were selected for the AI-based review, addressing key research questions. Further, Cochrane protocol was applied for risk-of-bias computations for selecting AI techniques. We studied the six types of software bug triaging techniques (SBTT) that were analyzed. AI has provided the possibility of automating the time-consuming manual SBT process. Our study shows that AI-based architectures, developers for newly reported bugs can be identified more accurately and quickly. Deep learning (DL)-based approaches demonstrate capabilities for developing SBT systems having improved (i) learning rate, (ii) scalability, and (iii) performance as compared to conventional approaches. For evaluating the SBT techniques, apart from the accuracy, precision, and recall, the mean average precision (mAP) is suggested to be an effective metric. In the future, more work is expected in the direction of SBT considering additional information from developer's networks, other repositories, and modern AI technologies.
•Designing medical devices involves various aspects of ergonomics.•CEREBO®, a noninvasive mNIRS device, allows early detection of intracranial hemorrhage in TBI.•Healthcare professionals from various ...backgrounds were evaluated for the study.•High usability was demonstrated thus promoting effective adoption in trauma care.
This study assessed machine learning powered Near-infrared spectroscopy based (mNIRS) device’s usability and human factor ergonomics in four distinct healthcare provider groups.
Traumatic Brain Injury (TBI) is a global concern with significant well-being implications. Timely intracranial hemorrhage (ICH) detection is crucial. mNIRS offers efficient non-invasive TBI screening.
Two device utilization stages involved operators (N = 21) and TBI-suspected subjects (n = 120). A hybrid approach used qualitative and quantitative methods, utilizing a 57-item survey and task completion time.
All groups positively perceived user-interface, physical, cognitive, and organizational ergonomics. The device's ease of use, calibration, size, cognitive support, and integration gained appreciation. Training reduced task completion time from 16.5 to 13.2 s.
mNIRS-based CEREBO® proves usable for TBI point-of-care assessment. Positive feedback from diverse healthcare groups validates design and cost-effectiveness alignment. A hybrid approach, training, and practice scans enhance usage and experience.
Fault management is an expensive process and analyzing data manually requires a lot of resources. Modern software bug tracking systems may be armed with automated bug report assignment functionality ...that facilitates bug classification or bug assignment to proper development group. For supporting decision systems, it would be beneficial to introduce information related to explainability. The purpose of this work is to evaluate the use of explainable artificial intelligence (XAI) in processes related to software development and bug classification based on bug reports created by either software testers or software users. The research was conducted on two different datasets. The first one is related to classification of security vs non-security bug reports. It comes from a telecommunication company which develops software and hardware solutions for mobile operators. The second dataset contains a list of software bugs taken from an opensource project. In this dataset the task is to classify issues with one of following labels crash, memory, performance, and security. Studies on XAI-related algorithms show that there are no major differences in the results of the algorithms used when comparing them with others. Therefore, not only the users can obtain results with possible explanations or experts can verify model or its part before introducing into production, but also it does not provide degradation of accuracy. Studies showed that it could be put into practice, but it has not been done so far.
The use of teleradiology for triaging of maxillofacial trauma Brucoli, Matteo; Boffano, Paolo; Franchi, Stefano ...
Journal of cranio-maxillo-facial surgery,
October 2019, 2019-Oct, 2019-10-00, 20191001, Letnik:
47, Številka:
10
Journal Article
Recenzirano
The aim of this study was to assess and discuss our experience with a teleradiology technique applied to facial trauma patients referred to an oral and maxillofacial surgery hub center.
All trauma ...patients with maxillofacial fractures from the hospitals of Vercelli, Biella, Borgosesia, Borgomanero, Verbania, and Domodossola who were referred between July 2014 and September 2018 to the hub maxillofacial center of Novara were reviewed. The following data were recorded for each patient: sex, age, referral hospital, etiology, etiology mechanisms, site of facial fractures, date of injury, indications for surgery according to teleradiology consultation, indications for surgery following clinical maxillofacial assessment, date of eventual surgery, timing of surgery from trauma, type of surgical intervention.
A total of 467 patients with a total of 605 fractures were triaged and managed by the Tempore telemedicine system. The most frequent cause of maxillofacial injury was fall. The most frequently observed fracture involved the zygoma. Following remote computed tomography assessment, surgical indications were suggested in 68 patients; 223 patients were not considered suitable candidates for surgery; and 176 patients needed a clinical assessment for the establishment of definitive eventual indications for surgery. Following clinical assessment, the absence and presence of surgical indications was confirmed in all 223 and 68 patients, respectively. Within the 176 patients with “possible” surgical indications, only 27 patients were referred for surgery.
Teleradiology may be helpful for an appropriate triaging of trauma patients from peripheral hospitals for the correct referral to a maxillofacial trauma hub center.
In previous work, we deployed IssueTAG, which uses the one-line summary and the description fields of the issue reports to automatically assign them to the stakeholders, who are responsible for ...resolving the reported issues. Since its deployment on
J
a
n
u
a
r
y
12, 2018 at Softtech – the software subsidiary of the largest private bank in Turkey, IssueTAG has made a total of 301,752 assignments (as of
N
o
v
e
m
b
e
r
2021). One observation we make is that a large fraction of the issue reports submitted to Softtech has screenshot attachments and, in the presence of such attachments, the reports often convey less information in their one-line summary and the description fields, which tends to reduce the assignment accuracy. In this work, we use the screenshot attachments as an additional source of information to further improve the assignment accuracy, which, to the best of our knowledge, has not been studied before for automatic issue assignments. In particular, we develop a number of multi-source assignment models, which use both the issue reports and the screenshot attachments, as well as a number of single source models, which use either the issue reports or the screenshot attachments, and empirically evaluate them on real issue reports. Compared to the currently deployed single-source model in the field, the best multi-source model improved the assignment accuracy from 0.848 to 0.855 at an acceptable overhead cost, reducing the overall 3.3 percentage-point deficit between the human triagers and the deployed system by 0.7 points.
We automate the process of assigning issue reports to development teams by using data mining approaches and share our experience gained by deploying the resulting system, called
IssueTAG
, at
...Softtech
. Being a subsidiary of the largest private bank in Turkey, Softtech on average receives 350 issue reports daily from the field, which need to be handled with utmost importance and urgency. IssueTAG has been making all the issue assignments at Softtech since its deployment on Jan 12, 2018. Deploying IssueTAG presented us not only with an unprecedented opportunity to observe the practical effects of automated issue assignment, but also with an opportunity to carry out user studies, both of which (to the best of our knowledge) have not been done before in this context. We first empirically determine the data mining approach to be used in IssueTAG. We then deploy IssueTAG and make a number of valuable observations. First, it is not just about deploying a system for automated issue assignment, but also about designing/changing the assignment process around the system. Second, the accuracy of the assignments does not have to be higher than that of manual assignments in order for the system to be useful. Third, deploying such a system requires the development of additional functionalities, such as creating human-readable explanations for the assignments and detecting deteriorations in assignment accuracies, for both of which we have developed and empirically evaluated different approaches. Last but not least, stakeholders do not necessarily resist change and gradual transition helps build confidence.
Purpose
Lower gastrointestinal bleeding (LGIB) is common and risk stratification scores can guide clinical decision-making. There is no robust risk stratification tool specific for LGIB, with ...existing tools not routinely adopted. We aimed to develop and validate a risk stratification tool for LGIB.
Methods
Retrospective review of LGIB admissions to three centres between 2010 and 2018 formed the derivation cohort. Using regressional analysis within a machine learning technique, risk factors for adverse outcomes were identified, forming a simple risk stratification score—The Birmingham Score. Retrospective review of an additional centre, not included in the derivation cohort, was performed to validate the score.
Results
Data from 469 patients were included in the derivation cohort and 180 in the validation cohort. Admission haemoglobin OR 1.07(95% CI 1.06–1.08) and male gender OR 2.29(95% CI 1.40–3.77) predicted adverse outcomes in the derivation cohort AUC 0.86(95% CI 0.82–0.90) which outperformed the Blatchford 0.81(95% CI 0.77–0.85), Rockall 0.60(95% CI 0.55–0.65) and AIM65 0.55(0.50–0.60) scores and in the validation cohort AUC 0.80(95% CI 0.73–0.87) which outperformed the Blatchford 0.77(95% CI 0.70–0.85), Rockall 0.67(95% CI 0.59–0.75) and AIM 65 scores 0.61(95% CI 0.53–0.69). The Birmingham Score also performs well at predicting adverse outcomes from diverticular bleeding AUC 0.87 (95% CI 0.75–0.98). A score of 7 predicts a 94% probability of adverse outcome.
Conclusion
The Birmingham Score represents a simple risk stratification score that can be used promptly on patients admitted with LGIB.
Software development is a modular approach involving multiple developers and multi-tasking teams working at different locations. A particular term in a software bug can belong to multiple modules and ...multiple developers’ profiles. Also, many people who report software bugs are unfamiliar with the exact technical terminology of software development, which causes the software bug to be unlabeled, vague, and noisy. Hence, analyzing, understanding, and assigning the newly reported bugs to the most appropriate developer is a challenging task for the triager. Intuitionistic Fuzzy Sets (IFS) consider the non-membership and hesitant values along with the membership values of the software bug terms mapped to the developers and thus provide a powerful tool for better analysis in cases where the same term can belong to multiple categories. Two IFS similarity measure-based techniques, namely, the Intuitionistic Fuzzy Similarity Model for Developer Term Relation (IFSDTR) and the Intuitionistic Fuzzy Similarity Model for Developer Category Relation (IFSDCR), are proposed in this work. In IFSDTR, a developer-term vocabulary is constructed based on the previous bug-fixing experience of software developers by considering the most frequent terms in the IFS representation of bugs they fixed earlier. In IFSDCR, software bugs are categorized into multiple categories and a developer-category relation is constructed. When a new bug is reported, the IFS similarity measure is calculated with the developer-term and developer-category relationship, and a fuzzy α-cut is applied to find a group of expert developers to fix it. The proposed techniques are evaluated on the available data set and compared with existing approaches to bug triaging. On the Eclipse, Mozilla, and NetBeans data sets, the IFSDTR techniques yield an accuracy of 0.90, 0.89, and 0.87, respectively, whereas the IFSDCR yields a greater accuracy of 0.93, 0.90, and 0.88 for the Eclipse, Mozilla, and NetBeans data sets, respectively. Similarly, in all other performance measures, the proposed approaches outperform the state-of-the-art approaches.
ABSTRACT
Objectives:
Diagnosis of patients suspected of mild dementia (MD) is a challenge and patient numbers continue to rise. A short test triaging patients in need of a neuropsychological ...assessment (NPA) is welcome. The Montreal cognitive assessment (MoCA) has high sensitivity at the original cutoff <26 for MD, but results in too many false-positive (FP) referrals in clinical practice (low specificity). A cutoff that finds all patients at high risk of MD without referring to many patients not (yet) in need of an NPA is needed. A difficulty is who is to be considered at risk, as definitions for disease (e.g. MD) do not always define health at the same time and thereby create subthreshold disorders.
Design:
In this study, we compared different selection strategies to efficiently identify patients in need of an NPA. Using the MoCA with a double threshold tackles the dilemma of increasing the specificity without decreasing the sensitivity and creates the opportunity to distinguish the clinical (MD) and subclinical (MCI) state and hence to get their appropriate policy.
Setting/participants:
Patients referred to old-age psychiatry suspected of cognitive impairment that could benefit from an NPA (
n
= 693).
Results:
The optimal strategy was a two-stage selection process using the MoCA with a double threshold as an add-on after initial assessment. By selecting who is likely to have dementia and should be assessed further (MoCA<21), who should be discharged (≥26), and who’s course should be monitored actively as they are at increased risk (21<26).
Conclusion:
By using two cutoffs, the clinical value of the MoCA improved for triaging. A double-threshold MoCA not only gave the best results; accuracy, PPV, NPV, and reducing FP referrals by 65%, still correctly triaging most MD patients. It also identified most MCIs whose intermediate state justifies active monitoring.