Cereal grains are an important part of human diet; hence, there is a need to maintain high quality and these grains must be free of physical and biological contaminants. A procedure was developed to differentiate physical contaminants from wheat using NIR (1000-1600 nm) hyperspectral imaging. Three experiments were conducted to select the best combinations of spectral pre-processing technique and statistical classifier to classify physical contaminants: seven foreign material types (barley, canola, maize, flaxseed, oats, rye, and soybean); six dockage types (broken wheat kernels, buckwheat, chaff, wheat spikelets, stones, and wild oats); and two animal excreta types (deer and rabbit droppings) from Canada Western Red Spring (CWRS) wheat. These spectra were processed using five spectral pre-processing techniques (first derivative, second derivative, Savitzky-Golay (SG) smoothing and differentiation, multiplicative scatter correction (MSC), and standard normal variate (SNV)). The raw and pre-processed data were classified using Support Vector Machines (SVM), Naïve Bayes (NB), and k-nearest neighbors (k-NN) classifiers. In each experiment, two-way and multi-way classifications were conducted. Among all the contaminant types, stones, chaff, deer droppings and rabbit droppings were classified with 100% accuracy using the raw reflectance spectra and different statistical classifiers. The SNV technique with k-NN classifier gave the highest accuracy for the classification of foreign material types from wheat (98.3±0.2%) and dockage types from wheat (98.9±0.2%). The MSC and SNV techniques with SVM or k-NN classifier gave perfect classification (100.0±0.0%) for the classification of animal excreta types from wheat. Hence, the SNV technique with k-NN classifier was selected as the best model. Two separate model performance evaluation experiments were conducted to identify and quantify (by number) the amount of contaminant type present in wheat. The overall identification accuracy of the first degree of contamination (one contaminant type with wheat) and the highest degree of contamination (all the contaminant type with wheat) was 97.6±1.6% and 92.5±6.5%, for foreign material types; 98.0±1.8% and 94.3±6.2%r for dockage types; and 100.0±0.0% and 100.0±0.0%, respectively for animal excreta types. The canola, stones, deer, and rabbit droppings were perfectly quantified (100.0±0.0%) at all the levels of contaminations. / February 2016

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第20054号 / 医博第4162号 / 新制||医||1018(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 湊谷 謙司, 教授 富樫 かおり, 教授 木村 剛 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Thesis (PhD)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: Maize (Zea mays L.) is used for human and animal consumption in diverse forms, from specialised foods in developed countries, to staple food in developing countries. Unfortunately, maize is prone to infection by different Fusarium species that can produce harmful mycotoxins. Fusarium verticillioides is capable of asymptomatic infection, where infected kernels show no sign of fungal growth, but are contaminated with mycotoxins. If fungal contamination is not detected early on, mycotoxins can enter the food chain. Rapid and accurate methods are required to detect, identify and distinguish between pathogens to enable swift decisions regarding the fate of a batch or consignment of cereal. Near infrared (NIR) hyperspectral imaging and multivariate image analysis (MIA) were evaluated to investigate the fungal development in maize kernels over time. When plotting principal component (PC) 4 against PC5, with percentages sum of squares (%SS) 0.49% and 0.34%, three distinct clusters were apparent in the score plot and this was associated with degree of infection. Prominent peaks at 1900 nm and 2136 nm confirmed that the source of variation was due to changes in starch and protein. Variable importance plots (VIP) confirmed the peaks observed in the PCA loading line plots. Early detection of fungal contamination and activity (20 h after inoculation) was possible before visual symptoms of infection appeared. Using NIR hyperspectral imaging and MIA it was possible to differentiate between species of Fusarium associated with maize. It was additionally applied to examine the fungal growth kinetics on culture media. Partial least squares discriminant analysis (PLS-DA) prediction results showed that it was possible to discriminate between species, with F. verticillioides the least correctly predicted (between 16-47% pixels correctly predicted). For F. subglutinans 78-100% and for F. proliferatum 60-80% pixels were correctly predicted. Three prominent bands at 1166, 1380 and 1918 nm were considered to be responsible for the differences between the growth zones. Variations in the bands at 1166 and 1380 nm were correlated with the depletion of carbohydrates as the fungus grew while the band at 1918 nm was a possible indication of spore and new mycelial formation. By plotting the pixels from the individual growth zones as a function of time, it was possible to visualise the emergence and interaction of the growth zones as separate growth profiles. The microstructure of fungal infected maize kernels was studied over time using high resolution X-ray micro-computed tomography (μCT). The presence of voids and airspaces could be seen in two dimensional (2D) X-ray transmission images and in the three dimensional (3D) tomograms. Clear differences were detected between kernels imaged after 20 and 596 h of inoculation. This difference in voids as the fungus progressed showed the effect of fungal damage on the microstructure of the maize kernels. Imaging techniques are important for rapid, accurate and objective evaluation of products for quality and safety. NIR hyperspectral imaging offers rapid chemical evaluation of samples in 2D images while μCT offers 3D microstructural information. By combining these image techniques more value was added and this led to a comprehensive evaluation of Fusarium infection in maize. / AFRIKAANSE OPSOMMING: Mielies (Zea mays L.) word in verskeie vorms deur mens en dier verbruik, van gespesialiseerde voedsel in ontwikkelde lande, tot stapelvoedsel in ontwikkelende lande. Ongelukkig is mielies onderhewig aan besmetting deur verskeie Fusarium spesies wat skadelike mikotoksiene kan produseer. Fusarium verticilloioides is in staat tot asimptomatiese infeksie waar die besmette pit geen teken van fungusgroei toon nie, maar (reeds) met mikotoksiene besmet is. Indien fungusbesmetting nie vroegtydig opgespoor word nie, kan mikotoksiene die voedselketting betree. Vinnige en akkurate metodes word benodig om patogene op te spoor, te identifiseer en ook om onderskeid tussen patogene te tref om sodoende (effektiewe) besluite aangaande die gebruik van ‘n lot of besending graan te neem. Naby-infrarooi (NIR) hiperspektrale beelding en meerveranderlike beeld ontleding (MIA) is geëvalueer om fungusontwikkeling in mieliepitte oor tyd te ondersoek. Wanneer hoofkomponent (PC) 4 teenoor PC5 gestip word, met persentasies som van kwadrate (%SS) 0.49% en 0/34%, is drie afsonderlike groepein die telling grafiek waargeneem. Dit is geassosieer met die graad van besmetting. Prominente pieke by 1900 nm en 2136 nm het bevestig dat veranderinge in stysel en proteïene die bron van die variasie was. Veranderlike belangrikheidsgrafieke (VIP) het die pieke wat in die PCA beladingslyngrafieke waargeneem is, bevestig. Vroegtydige opsporing (bespeuring) van fungusbesmetting en aktiwiteit (20 h na inokulasie) was moontlik voor visuele besmettingsimptome verskyn het. Onderskeid tussen Fusarium spesies wat met mielies geassosieer word, was moontlik deur gebruik te maak van NIR hiperspektrale beelding en MIA. Dit is bykomend toegepas om fungusgroeikinetika op kwekingsmedia te bestudeer. Parsiële kleinste kwadrate diskriminantanalise (PLS-DA) voorspellingsresultate het getoon dat dit moontlik was om tussen spesies te onderskei, met F. verticillioides die minste korrek voorspel (tussen 19-47% beeldelemente korrek voorspel). Vir F. subglutinans is 78-100% en vir F. proliferatum is 60-80% beeldelemente korrek voorspel. Drie prominente bande by 1166, 1380 en 1918 nm is oorweeg as oorsaak vir die verskille tussen die groeisones. Variasies in die bande by 1166 en 1380 nm is gekorreleer met die vermindering van koolhidrate soos die fungus groei, terwyl die band by 1918 nm ‘n moontlike aanduiding van spoor en nuwe miseliale vorming is. Deur die beeldelemente van die individuele groeisones as ‘n funksie van tyd te stip, was dit moontlik om die verskyning en interaksie van die groeisones as aparte groeiprofiele te visualiseer. Hoë-resolusie X-straal mikro-berekende tomografie (μCT) is gebruik om die mikrostruktuur van fungusbesmette mieliepitte oor tyd te ondersoek. Die voorkoms van leemtes en lugruimtes kon in die twee-dimensionele (2D) X-straal transmissie beelde en in die drie-dimensionele (3D) tomogramme gesien word. Duidelike verskille is waargeneem tussen pitte wat na 20 en 596 h na inokulasie verbeeld is. Hierdie verskil in leemtes soos die fungus vorder, het die effek van fungusskade op die mikrostruktuur van mieliepitte getoon. Beeldingstegnieke is belangrik vir vinnige, akkurate en objektiewe evaluasie van produkte vir kwaliteit en veiligheid. NIR hiperspektrale beelding bied vinnige chemiese evaluering van monsters in 2D beelde, terwyl μCT 3D mikrostrukturele inligting gee. Meer waarde is toegevoeg deur hierdie beeldingstegnieke te kombineer en dit het gelei tot ‘n omvangryke evaluering van Fusarium besmetting in mielies.

Thesis (MSc Food Sc)--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: The use of near infrared (NIR) hyperspectral imaging for distinguishing between pre-germinated and non pre-germinated barley, wheat and sorghum kernels and, the effect of kernel shape on hyperspectral images, have been investigated. Two sample sets were imaged. The first sample set was divided into six subsets; these subsets were treated with water and left to pre-germinate for different times (0, 6, 9, 12, 18 and 24 hrs). Subset viability was determined with the tetrazolium test. The second sample set was divided into seven subsets, treated with water and left to pre-germinate for 0, 3, 6, 9, 12, 18, 24 or 30 hrs. Individual kernel viability was determined with the tetrazolium test. NIR hyperspectral images were acquired using two different SisuCHEMA hyperspectral imaging systems. The first system acquired images with a 150 9m spatial resolution (first sample set) and the second system acquired images with a 30 9m spatial resolution (second sample set). Principal component analysis (PCA) was performed and a distinction between pre-germinated and non pre-germinated kernels was illustrated in PCA score images. Loading line plots showed that the main compounds contributing to spectral variation were starch, water and protein. These compounds were related to starch and protein hydrolysis. The distinction between pre-germinated and non pre-germinated kernels observed in the 30 9m spatial resolution images indicated NIR hyperspectral imaging was perhaps sensing incomplete endosperm degradation. Some kernels determined as pre-germinated by the tetrazolium test had the same chemical composition according to the score image as non pre-germinated kernels in the 30 9m spatial resolution images. A partial least squares discriminant analysis (PLS-DA) model with two classes (pre- germinated and non pre-germinated) was developed for each of the cultivars of the first sample set. The two classes were assigned in principal component (PC) 1 vs. PC 5 score plots. The model created for the barley cultivars resulted in excessive false positives and false negatives. The prediction results of wheat cultivars revealed that the model had a classification rate of 81% for the non pre-germinated class and 93% for the pre-germinated class. The sorghum prediction results revealed that the model correctly predicted 97% of the non pre-germinated class and 93% of the pre-germinated class. Two different PLS-DA models were developed for one image of each cultivar of the 30 9m spatial resolution images. The first model was developed by assigning each kernel in the score image and the second model was developed by assigning pixels in the score plot to either the pre- germinated or non pre-germinated class. Model 1 resulted in excessive false negatives. Model 2 resulted in excessive false positives. The differences between pre-germinated and non pre-germinated kernels were only observed in higher (PC 5 and 6) order PCs of the 150 9m spatial resolution images. The lower (PCs 1 to 4) order PCs (of each commodity) were subsequently examined with the aid of classification gradients. Kernel shape effects were observed in these PCs. The use of NIR hyperspectral imaging for distinguishing between pre-germinated and non pre-germinated grain kernels shows promise. / AFRIKAANSE OPSOMMING: Die gebruik van naby infrarooi (NIR) hiperspektrale beeld-analise is geëvalueer om onderskeid te tref tussen voor-ontkiemde en nie-voor-ontkiemde gars, koring en sorghum korrels. Die effek van korrelvorm op hiperspektrale beelde is ook geëvalueer. Die eerste stel graan-monsters is gebruik vir 150 9m ruimtelike resolusie beelde en die tweede stel is gebruik vir 30 9m ruimtelike resolusie beelde. Die eerste kultivar stel is verdeel in ses sub-stelle en met gedistilleerde water behandel vir 0, 6, 9, 12, 18 en 24 hr. Sub-stel lewensvatbaarheid is met die tetrazolium toets vasgestel. Elke kultivar in die tweede stel is in sewe sub-stelle verdeel en is vir 0, 3, 6, 9, 12, 18, 24 of 30 hr geïnkubeer. Individuele korrel lewensvatbaarheid is met die tetrazolium toets vasgestel. NIR hiperspektrale beelde is verkry deur gebruik te maak van twee verskillende SisuCHEMA kameras. Die verskillende kameras is gebruik om verskillende resolusie (30 en 150 9m ruimtelike resolusie) beelde te verkry. Hoofkomponent analise (HKA) is uitgevoer en ’n verskil tussen voor- ontkiemde en nie-voor-ontkiemde korrels is waargeneem in die 150 9m ruimtelike resolusie beelde. HK ladings stippe het water, stysel en proteïene uitgesonder as die verbindings wat bydrae het tot spektrale variasie. ’n Verskil tussen die voor-ontkiemde korrels en nie-voor-ontkiemde korrels is ook gesien vir die 30 9m ruimtelike resolusie beelde. Dit is egter ook waargeneem dat sommige korrels as voor-ontkiem bepaal is deur die tetrazolium toets, maar dié korrels het dieselfde chemiese samestelling volgens die punte beeld as nie-voor-ontkiemde korrels. Onvolledige endosperm hidrolise is ’n moontlike verduideliking vir die verskynsel. Die verbindings wat bygedra het tot die variasie is water, stysel en proteïene. ’n Parsiële kleinste kwadrate diskriminant analise (PKW-DA) model met twee klasse is ontwikkel vir elke kultivar van die 150 9m ruimtelike resolusie beelde. Die klasse is aangewys in the punte stip. Die model met die hoogste variasie in Y is gekies om die ander kultivars van dieselfde kommoditeit te voorspel. The PKW-DA resultate vir die gars kultivars het getoon dat die model vals positiewes en vals negatiewes opgelewer het. Die koring PKW-DA model het ’n klassifikasie koers van 81% vir die nie-voor-ontkiemde klasse en 93% vir die voor-ontkiemde klasse opgelewer. The PKW-DA resultate vir sorghum het getoon dat die model ’n klassifikasie koers van 97% vir die nie-voor-ontkiemde klasse en 93% vir die voor-ontkiemde klasse opgelewer. Twee verskillende PKW-DA modelle is ontwikkel vir elke beeld van elke kultivar van die 30 9m ruimtelike resolusie beelde. Die eerste model is ontwikkel deur elke korrel in die punte beeld aan te wys tot een van twee klasse en die tweede model is ontwikkel deur die beeldelemente in die punte stip tot een van twee klasse toe te skryf. Model 1 het vals negatiewes opgelewer en model 2 vals positiewes. Die verskille tussen die nie-voor-ontkiemde en voor-ontkiemde korrels is eers verduidelik in hoër orde HK van die 150 9m ruimtelike resolusie beelde. Die laer orde HK is dus ondersoek vir hul bydrae tot spektrale variasie met die hulp van klassifikasie gradiënte. Korrel vorm effekte is waargeneem. Die gebruik van NIR hiperspektrale beelding om onderskeid te tref tussen voor-ontkiemde en nie-voor-ontkiemde graan korrels, lyk belowend.

Thesis (MSc Food Sc)--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: Near infrared hyperspectral imaging (NIR HSI) in conjunction with multivariate image analysis was evaluated for the detection of millet and buckwheat flour in ground black pepper. Additionally, midinfrared (MIR) spectroscopy was used for the quantification of millet and buckwheat flour in ground black pepper. These techniques were applied as they allow non-destructive, invasive and rapid analysis. Black pepper and adulterant (either millet or buckwheat flour) mixtures were made in 5% (w/w) increments spanning the range 0-100% (w/w). The mixtures were transferred to eppendorf tube holders and imaged with a sisuChema short wave infrared (SWIR) pushbroom imaging system across the spectral range of 1000–2498 nm. Principal component analysis (PCA) was applied to pseudo-absorbance images for the removal of unwanted data (e.g. background, shading effects and bad pixels). PCA was subsequently applied to the ‘cleaned’ data. An adulterant concentration related gradient was observed in principal component one (PC1) and a difference between black pepper adulterated with buckwheat and millet was noted in PC4. Four absorption peaks (1461, 2241, 2303 and 2347 nm) were identified in the loading line plot of PC1 that are associated with protein and oil. The loading line plot of PC4 revealed absorption peaks at 1955, 1999, 2136 and 2303 nm, that are related to protein and oil. Partial least squares discriminant analysis (PLS-DA) was applied to NIR HSI images for discrimination between black pepper adulterated with varying amounts of adulterant (millet or buckwheat). The model created with millet adulterated black pepper samples had a classification accuracy of 77%; a classification accuracy of 70% was obtained for the buckwheat adulterated black pepper samples. An average spectrum was calculated for each sample in the NIR HSI images and the resultant spectra were used for the quantification of adulterant (millet or buckwheat) in ground black pepper. All samples were also analysed using an attenuated total reflectance (ATR) Fourier transform (FT) – infrared (IR) instrument and MIR spectra were collected between 576 and 3999 cm-1. PLS regression was employed. NIR based predictions (r2 = 0.99, RMSEP = 3.02% (w/w), PLS factor = 4) were more accurate than MIR based predictions (r2 = 0.56, RMSEP = 19.94% (w/w), PLS factors = 7). Preprocessed NIR spectra revealed adulterant specific absorption bands (1743, 2112 and 2167 nm) whereas preprocessed MIR spectra revealed a buckwheat specific signal at 1574 cm-1. NIR HSI has great promise for both the qualitative and quantitative analysis of powdered food products. Our study signals the beginning of incorporating hyperspectral imaging in the analysis of powdered food substances and results can be improved with advances in instrumental development and better sample preparation. / AFRIKAANSE OPSOMMING: Die gebruik van naby infrarooi hiperspektrale beelding (NIR HB) tesame met veelvoudige beeldanalise is ondersoek vir die opsporing van stysel-verwante produkte (giers en bokwiet) in gemaalde swart pepper. Middel-infrarooi (MIR) spektroskopie is addisioneel gebruik vir die kwantifisering van hierdie stysel-verwante produkte in swart pepper. Albei hierdie tegnieke is toegepas aangesien dit deurdringend van aard is en dit bied nie-destruktiewe sowel as spoedige analise. Swart pepper en vervalsingsmiddel (giers of bokwiet) mengsels is uitgevoer in 5% (m/m) inkremente tussen 0 en 100% (m/m). Eppendorfbuishouers is met die mengsels gevul en hiperspektrale beelde is verkry deur die gebruik van ‘n sisuChema SWIR (kortgolf infrarooi) kamera met ‘n spektrale reikwydte van 1000–2498 nm. Hoofkomponent-analise (HK) is toegepas op pseudo-absorbansie beelde vir die verwydering van ongewenste data (bv. agtergrond, skadu en dooie piksels). Hoofkomponent-analise is vervolgens toegepas op die ‘skoon’ data. Hoofkomponent (HK) een (HK1) het die aanwesigheid van ‘n vervalsingsmiddel konsentrasie verwante gradient getoon terwyl HK4 ‘n verskil getoon het tussen swart pepper vervals met giers en bokwiet. Vier absorpsiepieke (1461, 2241, 2303 en 2347 nm) was geïdentifiseer binne die HK lading stip van HK1 wat met proteïen en olie geassosieer kon word. Die HK lading stip van HK4 het absorpsipieke by 1955, 1999, 2136 en 2303 nm aangedui wat verband hou met proteïen en olie. Parsiële kleinste waarde diskriminant-analise (PKW-DA) is toegepas op die hiperspektrale beelde vir die moontlike onderskeiding tussen swart pepper vervals met verskeie hoeveelhede vervalsingsmiddel (giers of bokwiet). ‘n Klassifikasie koers van 77% is verkry vir die model ontwikkel met giers vervalsde swart pepper terwyl die model ontwikkel met bokwiet vervalsde swarte pepper ‘n klassifikasie koers van 70% bereik het. ‘n Gemiddelde spektrum is bereken vir elke monster in die hiperspektrale beelde en die resulterende spektra is gebruik vir die kwantifisering van vervalsingsmiddels (giers of bokwiet) in gemaalde swart pepper. ‘n ATR FT-IR instrument met spektrale reikwydte van 576-3999 cm-1 is additioneel gebruik vir die analise van alle monsters. Parsiële kleinste waarde regressie is gebruik vir kwantifikasie doeleindes. NIR gebasseerde voorspellings (r2 = 0.99, RMSEP = 3.02% (m/m), PLS faktore = 4) was meer akkuraat as die MIR gebasseerde voorspellings (r2 = 0.56, RMSEP = 19.94% (m/m), PLS faktore = 7). Vooraf behandelde NIR spektra het vervalsingsmiddel verwante absorpsiepieke (1743, 2112 en 2167 nm) aangetoon terwyl vooraf behandelde MIR spektra ‘n bokwiet verwante absorpsiepiek by 1574 cm-1 aangedui het. NIR HB toon goeie potensiaal vir beide kwalitatiewe en kwantitatiewe analise van gepoeierde voedsel produkte. Ons studie kan gesien word as die begin van die inkorporasie van hiperspektrale beelding in die analise van gepoeierde voedsel material en verbeterde resulte kan verkry word deur die vordering in instrumentasie ontwikkeling en verbeterde monstervoorbereiding.