Weed Detection Using Deep Learning: A Systematic Literature Review.

Weeds are one of the most harmful agricultural pests that have a significant impact on crops. Weeds are responsible for higher production costs due to crop waste and have a significant impact on the global agricultural economy. The importance of this problem has promoted the research community in exploring the use of technology to support farmers in the early detection of weeds. Artificial intelligence (AI) driven image analysis for weed detection and, in particular, machine learning (ML) and deep learning (DL) using images from crop fields have been widely used in the literature for detecting various types of weeds that grow alongside crops. In this paper, we present a systematic literature review (SLR) on current state-of-the-art DL techniques for weed detection. Our SLR identified a rapid growth in research related to weed detection using DL since 2015 and filtered 52 application papers and 8 survey papers for further analysis. The pooled results from these papers yielded 34 unique weed types detection, 16 image processing techniques, and 11 DL algorithms with 19 different variants of CNNs. Moreover, we include a literature survey on popular vanilla ML techniques (e.g., SVM, random forest) that have been widely used prior to the dominance of DL. Our study presents a detailed thematic analysis of ML/DL algorithms used for detecting the weed/crop and provides a unique contribution to the analysis and assessment of the performance of these ML/DL techniques. Our study also details the use of crops associated with weeds, such as sugar beet, which was one of the most commonly used crops in most papers for detecting various types of weeds. It also discusses the modality where RGB was most frequently used. Crop images were frequently captured using robots, drones, and cell phones. It also discusses algorithm accuracy, such as how SVM outperformed all machine learning algorithms in many cases, with the highest accuracy of 99 percent, and how CNN with its variants also performed well with the highest accuracy of 99 percent, with only VGGNet providing the lowest accuracy of 84 percent. Finally, the study will serve as a starting point for researchers who wish to undertake further research in this area.

A Nudge-Inspired AI-Driven Health Platform for Self-Management of Diabetes.

Diabetes mellitus is a serious chronic disease that affects the blood sugar levels in individuals, with current predictions estimating that nearly 578 million people will be affected by diabetes by 2030. Patients with type II diabetes usually follow a self-management regime as directed by a clinician to help regulate their blood glucose levels. Today, various technology solutions exist to support self-management; however, these solutions tend to be independently built, with little to no research or clinical grounding, which has resulted in poor uptake. In this paper, we propose, develop, and implement a nudge-inspired artificial intelligence (AI)-driven health platform for self-management of diabetes. The proposed platform has been co-designed with patients and clinicians, using the adapted 4-cycle design science research methodology (A4C-DSRM) model. The platform includes (a) a cross-platform mobile application for patients that incorporates a macronutrient detection algorithm for meal recognition and nudge-inspired meal logger, and (b) a web-based application for the clinician to support the self-management regime of patients. Further, the platform incorporates behavioral intervention techniques stemming from nudge theory that aim to support and encourage a sustained change in patient lifestyle. Application of the platform has been demonstrated through an illustrative case study via two exemplars. Further, a technical evaluation is conducted to understand the performance of the MDA to meet the personalization requirements of patients with type II diabetes.

Co-Design, Development, and Evaluation of a Mobile Solution to Improve Medication Adherence in Cancer: Design Science Research Approach.

BACKGROUND: Medication nonadherence negatively impacts the health outcomes of people with cancer as well as health care costs. Digital technologies present opportunities to address this health issue. However, there is limited evidence on how to develop digital interventions that meet the needs of people with cancer, are perceived as useful, and are potentially effective in improving medication adherence. OBJECTIVE: The objective of this study was to co-design, develop, and preliminarily evaluate an innovative mobile health solution called Safety and Adherence to Medication and Self-Care Advice in Oncology (SAMSON) to improve medication adherence among people with cancer. METHODS: Using the 4 cycles and 6 processes of design science research methodology, we co-designed and developed a medication adherence solution for people with cancer. First, we conducted a literature review on medication adherence in cancer and a systematic review of current interventions to address this issue. Behavioral science research was used to conceptualize the design features of SAMSON. Second, we conducted 2 design phases: prototype design and final feature design. Last, we conducted a mixed methods study on patients with hematological cancer over 6 weeks to evaluate the mobile solution. RESULTS: The developed mobile solution, consisting of a mobile app, a web portal, and a cloud-based database, includes 5 modules: medication reminder and acknowledgment, symptom assessment and management, reinforcement, patient profile, and reporting. The quantitative study (n=30) showed that SAMSON was easy to use (21/27, 78%). The app was engaging (18/27, 67%), informative, increased user interactions, and well organized (19/27, 70%). Most of the participants (21/27, 78%) commented that SAMSON's activities could help to improve their adherence to cancer treatments, and more than half of them (17/27, 63%) would recommend the app to their peers. The qualitative study (n=25) revealed that SAMSON was perceived as helpful in terms of reminding, supporting, and informing patients. Possible barriers to using SAMSON include the app glitches and users' technical inexperience. Further needs to refine the solution were also identified. Technical improvements and design enhancements will be incorporated into the subsequent iteration. CONCLUSIONS: This study demonstrates the successful application of behavioral science research and design science research methodology to design and develop a mobile solution for patients with cancer to be more adherent. The study also highlights the importance of applying rigorous methodologies in developing effective and patient-centered digital intervention solutions.

Development of a person-centred digital platform for the long-term support of people living with an adult-onset genetic disease predisposition: a mixed-methods study protocol.

INTRODUCTION: Individuals at an inherited high-risk of developing adult-onset disease, such as breast cancer, are rare in the population. These individuals require lifelong clinical, psychological and reproductive assistance. After a positive germline test result, clinical genetic services provide support and care coordination. However, ongoing systematic clinical follow-up programmes are uncommon. Digital health solutions offer efficient and sustainable ways to deliver affordable and equitable care. This paper outlines the codesign and development of a digital health platform to facilitate long-term clinical and psychological care, and foster self-efficacy in individuals with a genetic disease predisposition. METHODS AND ANALYSIS: We adopt a mixed-methods approach for data gathering and analysis. Data collection is in two phases. In phase 1, 300 individuals with a high-risk genetic predisposition to adult disease will undertake an online survey to assess their use of digital health applications (apps). In phase 2, we will conduct focus groups with 40 individuals with a genetic predisposition to cardiac or cancer syndromes, and 30 clinicians from diverse specialities involved in their care. These focus groups will inform the platform's content, functionality and user interface design, as well as identify the barriers and enablers to the adoption and retention of the platform by all endusers. The focus groups will be audiorecorded and transcribed, and thematic and content data analysis will be undertaken by adopting the Unified Theory of Acceptance and Use of Technology. Descriptive statistics will be calculated from the survey data. Phase 3 will identify the core skillsets for a novel digital health coordinator role. Outcomes from phases 1 and 2 will inform development of the digital platform, which will be user-tested and optimised in phase 4. ETHICS AND DISSEMINATION: This study was approved by the Peter MacCallum Human Research Ethics Committee (HREC/88892/PMCC). Results will be disseminated in academic forums, peer-reviewed publications and used to optimise clinical care.

Investigation of Intervention Solutions to Enhance Adherence to Oral Anticancer Medicines in Adults: Overview of Reviews.

BACKGROUND: Adherence to anticancer medicines is critical for the success of cancer treatments; however, nonadherence remains challenging, and there is limited evidence of interventions to improve adherence to medicines in patients with cancer. OBJECTIVE: This overview of reviews aimed to identify and summarize available reviews of interventions to improve adherence to oral anticancer medicines in adult cancer survivors. METHODS: A comprehensive search of 7 electronic databases was conducted by 2 reviewers who independently conducted the study selection, quality assessment using the A Measurement Tool to Assess Systematic Reviews 2, and data extraction. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 checklist was adapted to report the results. RESULTS: A total of 29 reviews were included in the narrative synthesis. The overall quality of the systematic reviews was low. The 4 main strategies to promote adherence were focused on education, reminders, behavior and monitoring, and multicomponent approaches. Digital technology-based interventions were reported in most reviews (27/29, 93%). A few interventions applied theories (10/29, 34%), design frameworks (2/29, 7%), or engaged stakeholders (1/29, 3%) in the development processes. The effectiveness of interventions was inconsistent between and within reviews. However, interventions using multiple strategies to promote adherence were more likely to be effective than single-strategy interventions (12/29, 41% reviews). Unidirectional communication (7/29, 24% reviews) and technology alone (11/29, 38% reviews) were not sufficient to demonstrate improvement in adherence outcomes. Nurses and pharmacists played a critical role in promoting patient adherence to oral cancer therapies, especially with the support of digital technologies (7/29, 24% reviews). CONCLUSIONS: Multicomponent interventions are potentially effective in promoting patient adherence to oral anticancer medicines. The seamless integration of digital solutions with direct clinical contacts is likely to be effective in promoting adherence. Future research for developing comprehensive digital adherence interventions should be evidence-based, theory-based, and rigorously evaluated.

Patient clustering using dynamic partitioning on correlated and uncertain biomedical data.

Background and objectivesHealth professionals look for specific patterns by correlating multiple physiological data in the process of deciding treatments to remedy clinical abnormalities. Biomedical data exhibit some common patterns in the event of identical clinical illnesses. The primary interest of this work is automatic discovery of such patterns in vital sign data (e.g. heart rate, blood pressure) using unsupervised learning and utilising them to identify patients with similar clinical conditions. MethodsA patient clustering method is developed that efficiently isolates patients into multiple groups by discovering dynamic patterns in multi-dimensional vital sign data. A dynamic partitioning algorithm and a patient clustering approach is proposed by introducing a measure namely aggregated instance-wise uncertainty (AIU) computed from multi-dimensional physiological time-series data. ResultsThe developed model is evaluated qualitatively using principal component analysis and silhouette value; and quantitatively in terms of its ability of clustering patients associated with different clinical situations. Experiments are conducted using real-world biomedical data of patients having various clinical conditions. Thee observed accuracy was 82.85% and 91.17% on two experimental datasets comprised of 35 and 34 patients data respectively.The comparisons show that the proposed approached outperformed than other methods in state-of-the-art approach. ConclusionsThe experimental outcomes demonstrate the effectiveness of the proposed approach in discovering distinct patterns with predictive significance.

A clinical decision-making mechanism for context-aware and patient-specific remote monitoring systems using the correlations of multiple vital signs.

BACKGROUND AND OBJECTIVES: In home-based context-aware monitoring patient's real-time data of multiple vital signs (e.g. heart rate, blood pressure) are continuously generated from wearable sensors. The changes in such vital parameters are highly correlated. They are also patient-centric and can be either recurrent or can fluctuate. The objective of this study is to develop an intelligent method for personalized monitoring and clinical decision support through early estimation of patient-specific vital sign values, and prediction of anomalies using the interrelation among multiple vital signs. METHODS: In this paper, multi-label classification algorithms are applied in classifier design to forecast these values and related abnormalities. We proposed a completely new approach of patient-specific vital sign prediction system using their correlations. The developed technique can guide healthcare professionals to make accurate clinical decisions. Moreover, our model can support many patients with various clinical conditions concurrently by utilizing the power of cloud computing technology. The developed method also reduces the rate of false predictions in remote monitoring centres. RESULTS: In the experimental settings, the statistical features and correlations of six vital signs are formulated as multi-label classification problem. Eight multi-label classification algorithms along with three fundamental machine learning algorithms are used and tested on a public dataset of 85 patients. Different multi-label classification evaluation measures such as Hamming score, F1-micro average, and accuracy are used for interpreting the prediction performance of patient-specific situation classifications. We achieved 90-95% Hamming score values across 24 classifier combinations for 85 different patients used in our experiment. The results are compared with single-label classifiers and without considering the correlations among the vitals. The comparisons show that multi-label method is the best technique for this problem domain. CONCLUSIONS: The evaluation results reveal that multi-label classification techniques using the correlations among multiple vitals are effective ways for early estimation of future values of those vitals. In context-aware remote monitoring this process can greatly help the doctors in quick diagnostic decision making.