Data-driven clustering : new methods and applications

Abstract : This work is concerned with the development and application of novel unsupervised learning methods, having in mind two target applications: the analysis of forensic case data and the classification of remote sensing images. First, a method based on a symbolic optimization of the inter-sample distance measure is proposed to improve the flexibility of spectral clustering algorithms, and applied to the problem of forensic case data. This distance is optimized using a loss function related to the preservation of neighborhood structure between the input space and the space of principal components, and solutions are found using genetic programming. Results are compared to a variety of state-of--the-art clustering algorithms. Subsequently, a new large-scale clustering method based on a joint optimization of feature extraction and classification is proposed and applied to various databases, including two hyperspectral remote sensing images. The algorithm makes uses of a functional model (e.g., a neural network) for clustering which is trained by stochastic gradient descent. Results indicate that such a technique can easily scale to huge databases, can avoid the so-called out-of-sample problem, and can compete with or even outperform existing clustering algorithms on both artificial data and real remote sensing images. This is verified on small databases as well as very large problems. Résumé : Ce travail de recherche porte sur le développement et l'application de méthodes d'apprentissage dites non supervisées. Les applications visées par ces méthodes sont l'analyse de données forensiques et la classification d'images hyperspectrales en télédétection. Dans un premier temps, une méthodologie de classification non supervisée fondée sur l'optimisation symbolique d'une mesure de distance inter-échantillons est proposée. Cette mesure est obtenue en optimisant une fonction de coût reliée à la préservation de la structure de voisinage d'un point entre l'espace des variables initiales et l'espace des composantes principales. Cette méthode est appliquée à l'analyse de données forensiques et comparée à un éventail de méthodes déjà existantes. En second lieu, une méthode fondée sur une optimisation conjointe des tâches de sélection de variables et de classification est implémentée dans un réseau de neurones et appliquée à diverses bases de données, dont deux images hyperspectrales. Le réseau de neurones est entraîné à l'aide d'un algorithme de gradient stochastique, ce qui rend cette technique applicable à des images de très haute résolution. Les résultats de l'application de cette dernière montrent que l'utilisation d'une telle technique permet de classifier de très grandes bases de données sans difficulté et donne des résultats avantageusement comparables aux méthodes existantes.

Extraction and concentration of vapors from fire debris for forensic purposes: evaluation of the use of Radiello Passive Air Sampler

The Radiello Passive Air Sampler is one of the latest innovations developed for the sampling of pollutants in the air by passive headspace. It has been reported that its properties allow an enhanced sensitivity, reproducibility and adsorption capacity. It therefore appears to be of interest in the extraction of potential residues of ignitable liquids present in fire debris when arson is suspected. A theoretical approach and several laboratory tests have made it possible to precisely characterize in a forensic perspective the potential of the device in extracting and concentrating the vapors of ignitable liquids found in fire debris. Despite some advantages, the Radiello device appears to be less efficient than traditional axial symmetry samplers.

Endoscopic extraction of a prevertebral migrated guidewire after posterior cervical instrumentation.

Images of Spine Care

Single-step extraction of fluconazole from plasma by ultra-filtration for the measurement of its free concentration by high performance liquid chromatography.

High performance liquid chromatography (HPLC) is the reference method for measuring concentrations of antimicrobials in blood. This technique requires careful sample preparation. Protocols using organic solvents and/or solid extraction phases are time consuming and entail several manipulations, which can lead to partial loss of the determined compound and increased analytical variability. Moreover, to obtain sufficient material for analysis, at least 1 ml of plasma is required. This constraint makes it difficult to determine drug levels when blood sample volumes are limited. However, drugs with low plasma-protein binding can be reliably extracted from plasma by ultra-filtration with a minimal loss due to the protein-bound fraction. This study validated a single-step ultra-filtration method for extracting fluconazole (FLC), a first-line antifungal agent with a weak plasma-protein binding, from plasma to determine its concentration by HPLC. Spiked FLC standards and unknowns were prepared in human and rat plasma. Samples (240 microl) were transferred into disposable microtube filtration units containing cellulose or polysulfone filters with a 5 kDa cut-off. After centrifugation for 60 min at 15000g, FLC concentrations were measured by direct injection of the filtrate into the HPLC. Using cellulose filters, low molecular weight proteins were eluted early in the chromatogram and well separated from FLC that eluted at 8.40 min as a sharp single peak. In contrast, with polysulfone filters several additional peaks interfering with the FLC peak were observed. Moreover, the FLC recovery using cellulose filters compared to polysulfone filters was higher and had a better reproducibility. Cellulose filters were therefore used for the subsequent validation procedure. The quantification limit was 0.195 mgl(-1). Standard curves with a quadratic regression coefficient > or = 0.9999 were obtained in the concentration range of 0.195-100 mgl(-1). The inter and intra-run accuracies and precisions over the clinically relevant concentration range, 1.875-60 mgl(-1), fell well within the +/-15% variation recommended by the current guidelines for the validation of analytical methods. Furthermore, no analytical interference was observed with commonly used antibiotics, antifungals, antivirals and immunosuppressive agents. Ultra-filtration of plasma with cellulose filters permits the extraction of FLC from small volumes (240 microl). The determination of FLC concentrations by HPLC after this single-step procedure is selective, precise and accurate.

Phosphorus saturation of Finnish soils : evaluating an easy oxalate extraction method

Selostus: Alumiini- ja rautaoksidien fosforikyllästysasteen arvioiminen suomalaisista peltomaista

DNA extraction using the QIAsymphony: Evaluation of PCR inhibitor removal

The goal of this study was to compare the quantity and purity of DNA extracted from biological tracesusing the QIAsymphony robot with that of the manual QIAamp DNA mini kit currently in use in ourlaboratory. We found that the DNA yield of robot was 1.6-3.5 times lower than that of the manualprotocol. This resulted in a loss of 8% and 29% of the alleles correctly scored when analyzing 1/400 and 1/800 diluted saliva samples, respectively. Specific tests showed that the QIAsymphony was at least 2-16times more efficient at removing PCR inhibitors. The higher purity of the DNA may therefore partlycompensate for the lower DNA yield obtained. No case of cross-contamination was observed amongsamples. After purification with the robot, DNA extracts can be automatically transferred in 96-wellsplates, which is an ideal format for subsequent RT-qPCR quantification and DNA amplification. Lesshands-on time and reduced risk of operational errors represent additional advantages of the robotic platform.

Transvaginal specimen extraction in colorectal surgery: current state of the art.

Aim  Background The expected benefit of transvaginal specimen extraction is reduced incision-related morbidity. Objectives A systematic review of transvaginal specimen extraction in colorectal surgery was carried out to assess this expectation. Method  Search strategy The following keywords, in various combinations, were searched: NOSE (natural orifices specimen extraction), colorectal, colon surgery, transvaginal, right hemicolectomy, left hemicolectomy, low anterior resection, sigmoidectomy, ileocaecal resection, proctocolectomy, colon cancer, sigmoid diverticulitis and inflammatory bowel diseases. Selection criteria Selection criteria included large bowel resection with transvaginal specimen extraction, laparoscopic approach, human studies and English language. Exclusion criteria were experimental studies and laparotomic approach or local excision. All articles published up to February 2011 were included. Results  Twenty-three articles (including a total of 130 patients) fulfilled the search criteria. The primary diagnosis was colorectal cancer in 51% (67) of patients, endometriosis in 46% (60) of patients and other conditions in the remaining patients. A concurrent gynaecological procedure was performed in 17% (22) of patients. One case of conversion to laparotomy was reported. In two patients, transvaginal extraction failed. In left- and right-sided resections, the rate of severe complications was 3.7% and 2%, respectively. Two significant complications, one of pelvic seroma and one of rectovaginal fistula, were likely to have been related to transvaginal extraction. The degree of follow up was specified in only one study. Harvested nodes and negative margins were adequate and reported in 70% of oncological cases. Conclusion  Vaginal extraction of a colorectal surgery specimen shows potential benefit, particularly when associated with a gynaecological procedure. Data from prospective randomized trials are needed to support the routine use of this technique.

Ex-PRESS R-50 miniature glaucoma implant insertion under the conjunctiva combined with cataract extraction

Rapport de synthèse : Le glaucome à angle ouvert est une neuropathie optique chronique progressive pour laquelle de nombreux traitements tant médicaux que chirurgicaux ont été proposés. La prise en charge chirurgicale s'articule principalement autour de deux chirurgies filtrantes, la trabéculectomie et la sclérectomie profonde avec implant de collagène. Cependant, les complications postopératoires de ces deux interventions étant relativement fréquentes, la recherche s'est orientée vers des traitements alternatifs dont la mise en place de micro-drains. Ces implants de drainage diminuent la pression intraoculaire en créant un court-circuit du flux d'humeur aqueuse de la chambre antérieure vers l'espace sous-conjonctival avec formation d'une bulle de filtration. L'implant Ex-PRESS R-50 est un implant miniature (2.5 mm de long pour 400 µm de diamètre) en acier inoxydable et biocompatible. La présente étude s'est proposée d'étudier l'efficacité et la sécurité de l'implant miniature Ex-Press R-50 lors d'une opération combinée cataracte-glaucome. Trente-cinq yeux de 35 patients (âge moyen: 75 ans) ont été inclus dans l'étude. Tous les patients ont bénéficié d'une opération de la cataracte par phacoemulsification et mise en place d'un implant de chambre postérieure suivie de l'implantation du micro-drain. Les pressions intraoculaires préopératoires et postopératoires, la meilleure acuité visuelle corrigée, le nombre de médicaments anti-glaucomateux ainsi que le type et le nombre de complications ont été évalués mensuellement puis tous les 6 mois pendant 4 ans. Le succès total a été défini par une pression postopératoire finale inférieure à 18mmHg sans traitement médical associé, le succès partiel par une pression postopératoire finale inférieure à 18mmHg avec ou sans traitement médical associé.. Le suivi moyen a été de 36.9 mois avec une baisse de la pression intraoculaire significative d'environ 25%. Une augmentation de l'acuité visuelle a été observée après l'opération de la cataracte et le nombre de médicaments anti-glaucomateux a été réduit de 57%. Dix patients ont bénéficié d'un traitement supplémentaire de la bulle de filtration par injection d'anti-métabolite (mitomycine C). Nous avons observé 8 complications majeures (4 érosions conjonctivales et 4 obstructions de l'orifice interne du micro-drain), toutes suivies de l'ablation de l'implant et de la réalisation d'une chirurgie classique du glaucome. En se basant sur les courbes de Kaplan-Meier à 48 mois, le taux de succès total était de 32.7% et le succès partiel de 53.7%. Nous pouvons conclure suite à ce travail que l'implant miniature Ex-PRESS R-50 est associé à un nombre trop élevé de complications, même si les cas non compliqués ont bénéficié d'une baisse significative de la pression intraoculaire. La modification de l'architecture du micro-drain ainsi que de la technique chirurgicale devrait augmenter le taux de succès.