Quantitative analysis of soil potassium by near-infrared (NIR) spectroscopy combined with a three-step progressive hybrid variable selection strategy.

Soil potassium is a crucial nutrient element necessary for crop growth, and its efficient measurement has become essential for developing rational fertilization plans and optimizing crop growth benefits. At present, data mining technology based on near-infrared (NIR) spectroscopy analysis has proven to be a powerful tool for real-time monitoring of soil potassium content. However, as technology and instruments improve, the curse of the dimensionality problem also increases accordingly. Therefore, it is urgent to develop efficient variable selection methods suitable for NIR spectroscopy analysis techniques. In this study, we proposed a three-step progressive hybrid variable selection strategy, which fully leveraged the respective strengths of several high-performance variable selection methods. By sequentially equipping synergy interval partial least squares (SiPLS), the random forest variable importance measurement (RF(VIM)), and the improved mean impact value algorithm (IMIV) into a fusion framework, a soil important potassium variable selection method was proposed, termed as SiPLS-RF(VIM)-IMIV. Finally, the optimized variables were fitted into a partial least squares (PLS) model. Experimental results demonstrated that the PLS model embedded with the hybrid strategy effectively improved the prediction performance while reducing the model complexity. The RMSET and RT on the test set were 0.01181% and 0.88246, respectively, better than the RMSET and RT of the full spectrum PLS, SiPLS, and SiPLS-RF(VIM) methods. This study demonstrated that the hybrid strategy established based on the combination of NIR spectroscopy data and the SiPLS-RF(VIM)-IMIV method could quantitatively analyze soil potassium content levels and potentially solve other issues of data-driven soil dynamic monitoring.

The burden of rheumatologic disease in Aboriginal and Torres Strait Islander Australians.

The objective of this article is to summarise the current knowledge regarding the prevalence of six rheumatological conditions in indigenous Australians - rheumatoid arthritis (RA), osteoarthritis (OA), osteoporosis (OSP), systemic lupus erythematosus (SLE), gout and musculoskeletal (MSK) pain. Online medical literature databases were searched for 'indigenous', 'Aboriginal' and 'Torres Strait Islander', as well as the names of the six conditions. Other included search terms were 'crystal', 'urate', 'arthritis' and 'arthropathy'. No limitations were placed on publication data or language. Forty-five articles examining the prevalence of the six conditions were identified. Based on the published literature, SLE appears to have a higher prevalence, while RA appears to have a lower prevalence in indigenous Australians compared to the non-indigenous community. MSK pain is prevalent, has a significant impact on indigenous people and is perceived as an important area of need. There is a paucity of data regarding these conditions in indigenous Australians. This may be impacted by the uncertainty of case ascertainment by self-report, differences in disease phenotypes and prevalence between the metropolitan compared to the rural or remote indigenous population, and difficulty with access to healthcare. Further studies in conjunction with local indigenous communities are needed to accurately determine the burden of rheumatological disease in the indigenous population. This will assist with resource and workforce planning to deliver culturally appropriate interventions. Strategies for future clinical work and research include the development and dissemination of culturally safe rheumatology resources, rheumatology training of Aboriginal Health Workers and wider integration of rheumatology clinics into community-controlled Aboriginal Health Services.

Cardiac sarcoma presenting as paraneoplastic arthritis and clubbing: a case report and literature review.

BACKGROUND: Cardiac tumours are rare, and clinical manifestations depend on the anatomical location. Symptoms can be the result of cardiac outflow anomalies, constitutional features such as fever, loss of weight, and/or paraneoplastic manifestations such as arthritis. To date, there has only been one other case report in the literature of cardiac sarcoma presenting as paraneoplastic arthropathy. CASE PRESENTATION: A 52-year-old woman presented with acute onset corticosteroid-resistant inflammatory polyarthralgia, clubbing and a systolic murmur. Transthoracic echocardiogram revealed a dilated left atrium with an echogenic mass and brain magnetic resonance imaging revealed multiple embolic infarcts. Histopathology following emergency resection showed a Grade 3 left atrial intimal sarcoma. The polyarthralgia and clubbing resolved soon after tumour removal. The patient went on to receive chemotherapy and remains in remission. CONCLUSIONS: This case highlights the rare paraneoplastic association of cardiac sarcoma and arthropathy.

A Novel Approach of Surface Texture Mapping for Cone-beam Computed Tomography in Image-guided Surgical Navigation.

The demand for cone-beam computed tomography (CBCT) imaging in clinics, particularly in dentistry, is rapidly increasing. Preoperative surgical planning is crucial to achieving desired treatment outcomes for imaging-guided surgical navigation. However, the lack of surface texture hinders effective communication between clinicians and patients, and the accuracy of superimposing a textured surface onto CBCT volume is limited by dissimilarity and registration based on facial features. To address these issues, this study presents a CBCT imaging system integrated with a monocular camera for reconstructing the texture surface by mapping it onto a 3D surface model created from CBCT images. The proposed method utilizes a geometric calibration tool for accurate mapping of the camera-visible surface with the mosaic texture. Additionally, a novel approach using 3D-2D feature mapping and surface parameterization technology is proposed for texture surface reconstruction. Experimental results, obtained from both real and simulation data, validate the effectiveness of the proposed approach with an error reduction to 0.32 mm and automated generation of integrated images. These findings demonstrate the robustness and high accuracy of our approach, improving the performance of texture mapping in CBCT imaging.

A decision tree network with semi-supervised entropy learning strategy for spectroscopy aided detection of blood hemoglobin.

Non-invasive techniques for rapid blood testing are gaining traction in global healthcare as they optimize medical screening, diagnosis and clinical decisions. Fourier transform infrared (FT-IR) spectroscopy is one of the most common technologies that can be used for non-destructive aided medical detection. Typically, after acquiring the Fourier transform infrared spectrum, spectral data preprocessing and feature extraction and quantitative analysis of several indicators of blood samples can be accomplished, in combination with chemometric method studies. At present, blood hemoglobin (HGB) concentration is one of the most valuable information for the clinical diagnosis of patient's health status. FT-IR spectroscopy is employed as a green technique aided medical test of blood HGB. Then the acquired HGB concentration data is switched to the spectral feature data by the studies of advanced chemometric method, in help for hiding the sensitive medical information to protect the privacy of patients. The decision tree network architecture is proposed for feature extraction of FT-IR data in order to find the small set of wavenumbers that are able to quantify HGB. A semi-supervised learning strategy is designed for tuning the number of network neuron nodes, in the way of searching for the maximum entropy increment. Each neuron is optimized by the growing of a semi-supervised decision tree, to accurately identify the informative FT-IR wavenumbers. The features extracted by the semi-supervised learning decision tree network guarantees the FT-IR aided detection model has high efficiency and high prediction accuracy. A model of quantifying the HGB concentration shows that the proposed decision tree network with semi-supervised entropy learning strategy outperforms the usual methods of full spectrum partial least square model and the fully connected neural network model in prediction accuracy. The framework is expected to support the FT-IR spectral technology for aided detection of medical and clinical data.

The disproportionately large contribution of the Maori and Pacific Islander community to the healthcare burden of gout in Western Sydney.

BACKGROUND: Gout is a common chronic inflammatory disorder due to monosodium urate deposition, which results in severe inflammatory arthritis. It is particularly common in those of Maori or Pacific Islander heritage. There is a significant number of this at-risk ethnic group in western Sydney. AIMS: To determine the healthcare burden of gout in Western Sydney. METHODS: We characterised patients managed in the emergency departments (EDs) of the four Western Sydney Local Health District (WSLHD) hospitals and those admitted for gout as the primary or secondary diagnosis from 1 January 2017 to 31 December 2018. RESULTS: There were 472 patients managed in ED on 552 occasions at a direct cost to the LHD of A$367 835. Those of Maori or Pacific Islander ethnicity comprised 25.2% (n = 119/472), while half (n = 39/80) of those managed in ED for gout on two or more occasions were of Maori or Pacific Islander ethnicity. Overall, 310 patients were admitted with gout as the principal diagnosis on 413 occasions at a cost of A$1.73 million. Seventy-five (24.2%) of the 310 patients were of Maori or Pacific Islander heritage. A total of 584 WSLHD inpatients had gout as a secondary diagnosis. This was associated with 714 admissions. CONCLUSIONS: The disproportionately large healthcare burden of gout in Western Sydney from the relatively small Maori and Pacific Islander population needs attention. Urgent culturally appropriate interventions to address gout are required to address this inequality.

Association between gout and cardiovascular outcomes in adults with no history of cardiovascular disease: large data linkage study in New Zealand.

Objective: To examine the association of gout with cardiovascular outcomes using linked administrative health data in Aotearoa New Zealand. Design: Data linkage study. Setting: National registries of pharmaceutical dispensing, hospital admission, and deaths linked to the Auckland/Northland regional repository of laboratory results to create a regional health contact population as of 31 December 2011. Participants: 942 416 residents of the Auckland/Northland region, aged 20-79 years with no history of cardiovascular disease. Main outcome measures: Time to first fatal or non-fatal cardiovascular event, identified from national datasets on hospital admissions and mortality, between 1 January 2012 and 31 December 2016. Cardiovascular disease was broadly defined as comprising ischaemic heart disease, ischaemic or haemorrhagic stroke, transient ischaemic attack, peripheral vascular disease, and heart failure. Interventions: A history of gout identified from a discharge diagnosis of gout from a public hospital admission or previous dispensing of gout specific drug treatments. The cohort was then linked to national hospital admissions and deaths through to 31 December 2016 (ie, 5 years' follow-up). Multivariable Cox proportional hazard models were constructed to assess the associations between gout, other risk factors, and cardiovascular outcomes. Results: Of 942 416 people included in the study, 31 907 (3.4%) had gout (6261 women and 25 646 men). After adjustment for multiple risk factors for cardiovascular disease, gout was associated with increased cardiovascular events (adjusted hazard ratio 1.34 (95% confidence interval 1.23 to 1.45) in women; 1.18 (1.12 to 1.24) in men). For men with gout, there was an increased risk of cardiovascular disease in those who were not dispensed regular allopurinol (1.15 (1.05 to 1.25)) and those with a serum urate above the treatment target of 0.36 mmol/L (1.16 (1.04 to 1.30)). Risk of cardiovascular events was lower for men with gout who were not dispensed colchicine compared with those who were (0.84 (0.77 to 0.92)). These findings were not observed in women. Conclusion: These results indicate that gout is associated with an increased risk of cardiovascular events. In men with gout without history of cardiovascular disease, the cardiovascular risk was lower in those regularly dispensed allopurinol and those with serum urate levels at the recommended treatment target. By contrast, colchicine dispensing was associated with an increased risk of cardiovascular events in men with gout without a cardiovascular history. The potential causal mechanisms of these associations require further exploration, including casual inference modelling in future studies.

Epstein-Barr virus-related lymphoma in rheumatoid arthritis: implications for long-term usage of immunosuppressive drugs and review of the literature.

BACKGROUND: Rheumatoid arthritis (RA) is a common autoimmune disease where methotrexate (MTX) is widely used as the first-line therapy. The combination of RA and MTX is associated with lymphoproliferative disorders (LPD). RA patients with Epstein-Barr virus (EBV) have impaired T-lymphocyte function, thus allowing an overgrowth of EBV-positive lymphoblastoid cells. We examined the association of EBV with LPD in immunosuppressed RA patients, particularly those treated with MTX. AIM: To review the relationship between RA, EBV-associated LPD and MTX use. METHODS: We reported two cases of RA patients with long-term MTX treatment who subsequently developed EBV-positive LPD, followed by a review of the relevant literature. RESULTS: Compared with normal population, RA patients have a higher risk of lymphoma, with diffuse large B-cell lymphoma being the most common subtype. MTX withdrawal can lead to lymphoma regression. Other biological therapies, such as abatacept and tocilizumab, are not associated with increased EBV-positive lymphoma diagnosis in RA patients. CONCLUSION: The association between EBV, lymphoma and MTX highlights the need to consider reducing or stopping MTX in patients who have had stable RA for many years.

Development of a radiographic scoring system for new bone formation in gout.

BACKGROUND: Features of new bone formation (NBF) are common in tophaceous gout. The aim of this project was to develop a plain radiographic scoring system for NBF in gout. METHODS: Informed by a literature review, scoring systems were tested in 80 individual 1st and 5th metatarsophalangeal joints. Plain radiography scores were compared with computed tomography (CT) measurements of the same joints. The best-performing scoring system was then tested in paired sets of hand and foot radiographs obtained over 2 years from an additional 25 patients. Inter-reader reproducibility was assessed using intraclass correlation coefficients (ICC). NBF scores were correlated with plain radiographic erosion scores (using the gout-modified Sharp-van der Heijde system). RESULTS: Following a series of structured reviews of plain radiographs and scoring exercises, a semi-quantitative scoring system for sclerosis and spur was developed. In the individual joint analysis, the inter-observer ICC (95% CI) was 0.84 (0.76-0.89) for sclerosis and 0.81 (0.72-0.87) for spur. Plain radiographic sclerosis and spur scores correlated with CT measurements (r = 0.65-0.74, P < 0.001 for all analyses). For the hand and foot radiograph sets, the inter-observer ICC (95% CI) was 0.94 (0.90-0.98) for sclerosis score and 0.76 (0.65-0.84) for spur score. Sclerosis and spur scores correlated highly with plain radiographic erosion scores (r = 0.87 and 0.71 respectively), but not with change in erosion scores over 2 years (r = -0.04-0.15). CONCLUSION: A semi-quantitative plain radiographic scoring method for the assessment of NBF in gout is feasible, valid, and reproducible. This method may facilitate consistent measurement of NBF in gout.

Review: Outcome measures in calcium pyrophosphate deposition.

This review highlights outcomes for patients with calcium pyrophosphate deposition (CPPD) reported in prior studies and underscores challenges to assessing outcomes of this condition. Prior clinical studies of interventions for CPPD focused on joint damage and calcification on imaging tests, joint pain, swelling, and inflammatory biomarkers. Qualitative interviews with patients with CPPD and healthcare providers additionally identified flares, overall function, and use of analgesic medications as important outcomes. Imaging evidence of joint damage and calcification is likely to be outcomes in future clinical studies of CPPD, though reliability and sensitivity to change in CPPD require further testing for several imaging modalities. Challenges to outcome measurement in CPPD include questions of attribution of signs and symptoms to CPPD versus co-existing forms of arthritis, lack of therapies to prevent or dissolve calcium pyrophosphate crystal deposition, absence of validated patient- or physician-reported CPPD outcome measures, and scarcity of large cohorts in which to study outcomes of different clinical presentations of CPPD.

Towards development of core domain sets for short term and long term studies of calcium pyrophosphate crystal deposition (CPPD) disease: A framework paper by the OMERACT CPPD working group.

INTRODUCTION: Although calcium pyrophosphate deposition (CPPD) is common, there are no published outcome domains or validated measurement instruments for CPPD studies. In this paper, we describe the framework for development of the Outcome Measures in Rheumatology (OMERACT) CPPD Core Domain Sets. METHODS: The OMERACT CPPD working group performed a scoping literature review and qualitative interview study. Generated outcomes were presented at the 2020 OMERACT CPPD virtual Special Interest Group (SIG) meeting with discussion focused on whether different core domain sets should be developed for different calcium pyrophosphate deposition (CPPD) clinical presentations and how the future CPPD Core Domain Set may overlap with already established osteoarthritis (OA) domains. These discussions informed development of a future work plan for development of the OMERACT CPPD Core Domain Sets. FINDINGS: Domains identified from a scoping review of 112 studies and a qualitative interview study of 36 people (28 patients with CPPD, 7 health care professionals, one stakeholder) were mapped to core areas of OMERACT Filter 2.1. The majority of SIG participants agreed there was need to develop separate core domain sets for "short term" and "long term" studies of CPPD. Although CPPD + OA is common and core domain sets for OA have been established, participants agreed that existing OA core domain sets should not influence the development of OMERACT core domain sets for CPPD. Prioritization exercises (using Delphi methodology) will consider 40 potential domains for short term studies of CPPD and 47 potential domains for long term studies of CPPD. CONCLUSION: Separate OMERACT CPPD Core Domain Sets will be developed for "short term" studies for an individual flare of acute CPP crystal arthritis and for "long term" studies that may include participants with any clinical presentation of CPPD (acute CPP crystal arthritis, chronic CPP crystal inflammatory arthritis, and/or CPPD + OA).

Experience and impact of crystal pyrophosphate deposition (CPPD) from a patient and caregiver perspective: A qualitative exploration from the OMERACT CPPD working group.

OBJECTIVE: To explore the lived experience of people with calcium pyrophosphate deposition (CPPD) disease and the impact of this condition on their daily lives. METHODS: Patients with CPPD and their caregivers were invited to take part in a one-to-one (patient only) or paired (patient and caregiver) semi-structured interview. Interviews covered patients' diagnosis and treatment experiences, and the impact of CPPD on their daily lives. Transcribed interviews were analysed using inductive thematic analysis. RESULTS: 28 patient interviews, six of which included a caregiver, were conducted across five countries. Acute CPP crystal arthritis flares resulted in temporary but profound disability for most patients, disrupting their ability to go about day-to-day activities, and they sought immediate medical attention. CPPD+OA and chronic CPP crystal inflammatory arthritis presented patients with longer term limitations in daily lives. Patients and their caregivers described these disruptions and limitations, which included a reduced ability or inability to complete household and self-care tasks, exercise, socialise, work and drive. They also described how arthritis pain and resulting limitations adversely impacted upon patients' psychological wellbeing. Delays in referral to specialists and diagnostic uncertainty were described by many. Lack of appropriate treatment or access to treatments only upon worsening of symptoms impacted upon the length of time some patients spent in pain and with functional limitations. CONCLUSION: This study is the first to demonstrate the wide-ranging impact of CPPD, and highlights the need for improved diagnosis, physician training, as well as greater emphasis upon finding targeted therapies to specifically treat CPPD.

Outcome domains reported by patients, caregivers, healthcare professionals and stakeholders for calcium pyrophosphate deposition (CPPD): A content analysis based on semi-structured qualitative interviews from the OMERACT CPPD working group.

INTRODUCTION: Although calcium pyrophosphate deposition (CPPD) disease is common, there are no validated outcome measures for clinical research in this condition. The aim of this study was to generate a list of outcome domains as reported by patients, their caregivers, healthcare professionals (HCPs) and stakeholders to inform the development of an Outcome Measures in Rheumatology (OMERACT) Core Domain Set for CPPD. METHODS: Patients with CPPD and their caregivers, HCPs and stakeholders took part in semi-structured qualitative interviews to explore potential outcome domains for CPPD clinical research relevant to their lived experience and knowledge of CPPD. Interviews were conducted in six countries across three continents. Data was analysed using manifest content analysis to identify outcome domains, which were tabulated and mapped to the core areas as defined by the OMERACT Filter 2.1. RESULTS: Thirty-six interviews were conducted in total. Participants comprised of 28 patients (six of which included a caregiver), seven HCPs and one stakeholder. The commonly identified (sub-) domains (d) across the 1) abnormalities/manifestations core area were joint pain (d = 35), joint swelling (d = 27), joint stiffness (d = 25), CPPD flares (d = 25); 2) life-impact core area were overall function (d=35), and specifically the ability to complete daily tasks (d = 25); and 3) societal/resource use core area were use of analgesic medicines (d = 26). Patients more commonly reported joint swelling, stiffness and range of movement, and use of analgesics while HCPs more commonly reported domains relating to presence of CPP crystals, radiologic calcification, joint damage, time to diagnosis and suitability of treatment. CONCLUSION: Among a number of potential outcome domains identified, articular manifestations, function and analgesic use were most frequently mentioned by participants. These findings will be used to develop an OMERACT Core Domain Set for CPPD.

A Novel Genetic Algorithm-Based Optimization Framework for the Improvement of Near-Infrared Quantitative Calibration Models.

The global fishmeal production is used for animal feed, and protein is the main component that provides nutrition to animals. In order to monitor and control the nutrition supply to animal husbandry, near-infrared (NIR) technology was utilized for rapid detection of protein contents in fishmeal samples. The aim of the NIR quantitative calibration is to enhance the model prediction ability, where the study of chemometric algorithms is inevitably on demand. In this work, a novel optimization framework of GSMW-LPC-GA was constructed for NIR calibration. In the framework, some informative NIR wavebands were selected by grid search moving window (GSMW) strategy, and then the variables/wavelengths in the waveband were transformed to latent principal components (LPCs) as the inputs for genetic algorithm (GA) optimization. GA operates in iterations as implementation for the secondary optimization of NIR wavebands. In steps of the variable's population evolution, the parametric scaling mode was investigated for the optimal determination of the crossover probability and the mutation operator. With the GSMW-LPC-GA framework, the NIR prediction effect on fishmeal protein was experimentally better than the effect by simply adopting the moving window calibration model. The results demonstrate that the proposed framework is suitable for NIR quantitative determination of fishmeal protein. GA was eventually regarded as an implementable method providing an efficient strategy for improving the performance of NIR calibration models. The framework is expected to provide an efficient strategy for analyzing some unknown changes and influence of various fertilizers.

Outcome domains reported in calcium pyrophosphate deposition studies: A scoping review by the OMERACT CPPD working group.

INTRODUCTION: Although calcium pyrophosphate deposition (CPPD) is common, there are no validated outcome domains and/or measurements for CPPD studies. The aim of this work was to identify domains that have been reported in prior clinical studies in CPPD, to inform the development of a core set of domains for CPPD studies. METHODS: We performed a scoping literature review for clinical studies in CPPD, searching in Medline (via PubMed), EMBASE, and Cochrane Central Register of Controlled Trials (CENTRAL) databases; published from January 1, 1946 to January 7, 2020. All reported outcomes and study design data were extracted and mapped to the core areas and domains as defined by the OMERACT Filter 2.1.The protocol was registered on PROSPERO (CRD: 42019137075; 09-07-2019). FINDINGS: There were 112 papers identified, comprising of 109 observational studies and three randomized controlled trials. Most studies reported clinical presentations of OA with CPPD or acute CPP crystal arthritis. Outcomes that mapped to 22 domains were identified; the most frequently reported measures mapped to the following domains/sub-domains: imaging (joint damage on imaging tests - 59 studies; joint calcification on imaging tests - 28 studies), joint pain (26 studies), response to treatment (23 studies), side effects of treatment (15 studies), inflammation in the joint fluid or blood (ESR or C-reactive protein - 12 studies; synovial fluid markers - 4 studies; other blood markers - 2 studies), overall function (14 studies), joint swelling (12 studies) and range of joint movement (10 studies). Very few studies mapped to domains related to life impact, societal/resource use or longevity. CONCLUSION: There is substantial variability in outcomes reported in CPPD studies. Outcomes that map to imaging manifestations, joint pain and response to treatment domains are most often reported.

Optimization Model for the Distribution of Fiducial Markers in Liver Intervention.

The distribution of fiducial markers is one of the main factors affected the accuracy of optical navigation system. However, many studies have been focused on improving the fiducial registration accuracy or the target registration accuracy, but few solutions involve optimization model for the distribution of fiducial markers. In this paper, we propose an optimization model for the distribution of fiducial markers to improve the optical navigation accuracy. The strategy of optimization model is reducing the distribution from three dimensional to two dimensional to obtain the 2D optimal distribution by using optimization algorithm in terms of the marker number and the expectation equation of target registration error (TRE), and then extend the 2D optimal distribution in two dimensional to three dimensional to calculate the optimal distribution according to the distance parameter and the expectation equation of TRE. The results of the experiments show that the averaged TRE for the human phantom is approximately 1.00 mm by applying the proposed optimization model, and the averaged TRE for the abdominal phantom is 0.59 mm. The experimental results of liver simulator model and ex-vivo porcine liver model show that the proposed optimization model can be effectively applied in liver intervention.

Kernel functions embedded in support vector machine learning models for rapid water pollution assessment via near-infrared spectroscopy.

Water pollution is a challenging problem encountered in total environmental development. Near-infrared (NIR) spectroscopy is a well-refined technology for rapid water pollution detection. Calibration models are established and optimized to search for chemometric algorithms with considerably improved prediction effects. Machine learning improves the prediction capability of NIR spectroscopy for the accurate assessment of water pollution. Least squares support vector machine (LSSVM) algorithm fits parameters to target problems in a data-driven manner. The modeling capability of this algorithm mainly depends on its kernel functions. In this study, the LSSVM method was used to establish NIR calibration models for the quantitative determination of chemical oxygen demand, which is a critical indicator of water pollution level. The effects of different kernels embedded in LSSVM were investigated. A novel kernel was proposed by using a logistic-based neural network. In contrast to common kernels, this novel kernel can utilize a deep learning approach for parameter optimization. The proposed kernel also strengthens model resistance to over-fitting such that cross-validation can be reasonably utilized. The proposed novel kernel is applicable for the quantitative determination of water pollution and is a prospective solution to other problems in the field of water resource management.

Rapid Detection of Pomelo Fruit Quality Using Near-Infrared Hyperspectral Imaging Combined With Chemometric Methods.

Pomelo is an important agricultural product in southern China. Near-infrared hyperspectral imaging (NIRHI) technology is applied to the rapid detection of pomelo fruit quality. Advanced chemometric methods have been investigated for the optimization of the NIRHI spectral calibration model. The partial least squares (PLS) method is improved for non-linear regression by combining it with the kernel Gaussian radial basis function (RBF). In this study, the core parameters of the PLS latent variables and the RBF kernel width were designed for grid search selection to observe the minimum prediction error and a relatively high correlation coefficient. A deep learning architecture was proposed for the parametric scaling optimization of the RBF-PLS modeling process for NIRHI data in the spectral dimension. The RBF-PLS models were established for the quantitative prediction of the sugar (SU), vitamin C (VC), and organic acid (OA) contents in pomelo samples. Experimental results showed that the proposed RBF-PLS method performed well in the parameter deep search progress for the prediction of the target contents. The predictive errors for model training were 1.076% for SU, 41.381 mg/kg for VC, and 1.136 g/kg for OA, which were under 15% of their reference chemical measurements. The corresponding model testing results were acceptably good. Therefore, the NIRHI technology combined with the study of chemometric methods is applicable for the rapid quantitative detection of pomelo fruit quality, and the proposed algorithmic framework may be promoted for the detection of other agricultural products.