Multi-class Method Research Articles

A MULTICLASS METHOD FOR THE ANALYSIS OF ANTIBIOTIC RESIDUES IN MILK POWDER AND WHEY POWDER BY UPLC-MS/MS

An ultra-performance liquid chromatography–tandem mass spectrometry (UPLC-MS/MS) multi-class method was developed and validated for the determination of 27 antibiotic residues from five classes (macrolides, sulfonamids, quinolones, lincosamids, diaminopyrimidines) in milk powder and whey powder which are commonly used as raw material of many products in the food industry. A simple, inexpensive and fast sample preparation based on the extraction of samples with methanol is the main advantage of the validated method. The validation of the method was performed following the requirements outlined in the European Commission Decision 2002/657/EC, which is required procedure for the methods used in analysis of official audits samples in Europe and Turkey. Matrix effect and uncertainties related with sample preparation and instrumental analysis were minimised by using matrix-matched calibration. The method is capable of detecting residues at very low concentrations.

Study on the Relationship between Working Ability and Working Performance of Health Personnel in Township Health Centers

Objective: To describe the status quo and influencing factors of the work ability and work performance of health personnel in township health centers, and to explore the relationship mechanism between them. Methods: Using a multi-class stratified sampling method, semi-structured personal in-depth interviews were conducted with 182 health personnel in 45 township health centers, word frequency analysis was performed on relevant content, and a framework of mediating effects of work motivation was constructed to analyze the effect of work ability on work performance. Results: The main factors to improve the working ability of health personnel are education and training, and the main factors that restrict working ability are external conditions and personnel qualification restrictions; among the factors that affect work performance, human resources are the primary human factor, income level is the primary economic factors and interpersonal relationship is the primary environmental factor. Conclusion: Continuing education, self and social identity have a positive impact on the improvement of health personnel’s work ability and behavioral performance, economic factors have a negative impact on health personnel’s satisfaction, and the new medical reform policy has both positive and negative impact on it. Work motivation produces a mediating effect between the work ability and work performance of health workers.

Validation of a multi-residue UHPLC-HRMS method for antibiotics screening in milk, fresh cheese, and whey

The aim of this study was to develop a UHPLC–HRMS methodology as a suitable tool for the screening of antibiotics in milk, fresh cheese, and whey. Compounds were analysed using an Orbitrap Exactive™ analyser working at a resolving power of 50,000 FWHM in full scan, both in positive and negative electrospray ionization mode, and a high mass accuracy (<5 ppm) and a retention time deviation of ±2.5 % were set as identification criteria. The multiclass method was developed for the simultaneous determination of 36 substances from different antibiotic groups (β-lactams, lincosamides, macrolides, quinolones, sulfonamides, and tetracyclines) and the quantification was performed by using matrix-matched calibration curves corrected by the corresponding internal standards. Matrix effect and recovery percentages were also calculated prior to the selection of internal standard. After comparing different sample preparation procedures, a quick and easy extraction with acetate buffer at pH 5.2 and acetonitrile (20/80, v/v), followed by C18 dispersive Solid Phase Extraction (dSPE) bulk sorbent was selected. The detection capabilities (CCβs) were established at 25 % or 50 % of the Maximum Residue Limit (MRL) set in milk (no regulation for cheese and whey), and precision (repeatability and reproducibility) and trueness (recovery) were satisfactory according to Commission Decision 2002/657/EC.

Multiclass cyanotoxin analysis in reservoir waters: Tandem solid-phase extraction followed by zwitterionic hydrophilic interaction liquid chromatography-mass spectrometry

The presence of cyanobacteria and cyanotoxins in all water bodies, including ocean water and fresh water sources, represents a risk for human health as eutrophication and climate change are enhancing their level of proliferation. For risk assessment and studies on occurrence, the development of reliable and sensitive analytical approaches able to cover a wide range of cyanotoxins is essential. This work describes the development of an HILIC-MS/MS multiclass method for the simultaneous analysis of eight cyanotoxins in reservoir water samples belonging to three different classes according to their chemical structure: cyclic peptides (microcystin-LR, microcystin-RR and nodularin), alkaloids (cylindrospermopsin, anatoxin-a) and three non-protein amino acids isomers such as β-methylamino-L-alanine, 2,4-diaminobutyric acid and N-(2-aminoethyl)glycine). A SeQuant ZIC-HILIC column was employed to achieve the chromatographic separation in less than 12 min. Previously, a novel sample treatment based on a tandem solid-phase extraction (SPE) system using mixed cation exchange (MCX) and Strata-X cartridges was investigated with the aim of extracting and preconcentrating this chemically diverse group of cyanotoxins. The Strata-X cartridge, which was configured first in the line of sample flow, retained the low polar compounds and the MCX cartridge, which was at the bottom of the dual system, retained mainly the non-protein amino acids. The optimization procedure highlighted the importance of sample ion content for the recoveries of some analytes such as the isomers β-N-methylamino-L-alanine and 2-4-diaminobutyric acid. Method validation was carried out in terms of linearity, limit of detection (LOD) and quantification (LOQ), recoveries, matrix effect and precision in terms of repeatability and intermediate precision. This work represents the first analytical method for the simultaneous analysis of these multiclass cyanotoxins in reservoir water samples, achieving LOQs in the very low range of 7·10−3 – 0.1 μg L−1. Despite high recoveries obtained at the LOQ concentration levels (101.0–70.9%), relative standard deviations lower than 17.5% were achieved.

Multi–class analysis of 30 antimicrobial residues in poultry feathers by liquid chromatography tandem mass spectrometry

ABSTRACT Poultry feathers are nowadays partially re-introduced into the animal food chain and the environment. They are valorised by their transformation into feather meal in order to be used as fertilisers in agriculture but also in animal feed (in particular, pet food and fish feed). However, unlike food producing animals for humans, feathers from poultry animals are not subject to a ban or regulatory limits on the presence of antibiotic residue after veterinary treatment. Feathers could therefore be a potential reservoir of antibiotic residues, unintentionally exposing the environment and animals through food, which might contribute to the emergence of antibiotic resistance. To this end, a multi-class liquid chromatographic method coupled with tandem mass spectrometry (LC-MS/MS) was developed for the detection and determination of residues of 30 antibiotics from eight groups of antibacterial (quinolones, lincosamides, macrolides, penicillins, phenicols, tetracyclines, sulphonamides and diaminopyrimidines) in feathers. The extraction of the analytes from the feathers was carried out by the salting out technique. The separation of the analytes employed a Kinetex C18 column. Quantification was made using internal standards. All analytes have been validated according to the performance criteria of Decision 2002/657/EC. Trueness of the method ranged from to 93% to 111% for all analytes and intermediate precision were to 1.2–18.8%. The limits of quantification (LOQ) were from 13 to 150 µg kg−1 depending on the analytes. The method is suitable for the monitoring and quantification of antibiotic residues in feathers over the range 13–600 µg kg−1 depending on the compound.

A single method to analyse residues from five different classes of prohibited pharmacologically active substances in milk

ABSTRACT In the European Union, the use of veterinary drugs belonging to the A6 group is prohibited in food-producing animals according to Commission Regulation (EU) No. 2010/37. The aim of this study was to improve the analytical control strategy by developing a single method to analyse residues of prohibited pharmacologically active substances in milk. For this, a single method was developed to analyse 16 prohibited pharmacologically active substances belonging to five different substance classes at required or recommended levels: nitroimidazoles at 3 μg kg−1, nitrofurans at 0.5 μg kg−1, chloramphenicol at 0.1 μg kg−1, dapsone at 5 μg kg−1 and chlorpromazine at 1 μg kg−1. Milk sample preparation started with an acid hydrolysis combined with a derivatisation. These steps were followed by a clean-up consisting of a dispersive solid-phase extraction and a liquid–liquid extraction. Finally, the sample extracts were analysed by liquid chromatography combined with tandem mass spectrometry, operating alternately in the positive and negative mode. The method was fully validated according to Commission Decision 2002/657/EC for bovine milk and additionally validated for caprine milk. The validation proved that the method is highly effective to detect and confirm all 16 substances in bovine and caprine milk and, additionally to quantify 15 of these substances in bovine milk and 13 of these substances in caprine milk. This study resulted in a new multi-class method to detect, quantify and confirm the identity of 16 prohibited pharmacologically active substances belonging to five different substance classes in two types of milk.

Development and validation of a multiclass confirmatory method for the determination of over 60 antibiotics in eggs using liquid-chromatography high-resolution mass spectrometry

A multiclass method for the determination of antimicrobial substances in eggs has been developed and validated, covering sixty-three antibiotics belonging to ten families. After extraction with acidified acetonitrile and EDTA, the analytes were injected into a liquid-chromatography high-resolution mass-spectrometry system operating in positive ionization mode. The procedure was successfully validated as confirmatory method evaluating selectivity, linearity, precision, recovery, decision limit, detection capability, limits of detection and quantitation. The here developed is the first method including all the antibiotics with Maximum Residue Limit (MRL) in eggs, except very polar drugs (aminoglycosides and colistins), which require specific protocols. Finally, a wide survey was carried out analyzing 100 commercial eggs samples from local markets. Only 3% contained residues of authorized substances at concentrations lower than the relevant MRLs.

A rapid multiclass method for antibiotic residues in goat dairy products by UPLC-quadrupole/electrostatic field orbitrap high-resolution mass spectrometry

BackgroundSulfanilamides, quinolones, nitroimidazoles, tetracyclines, cephalosporins, macrolides, and β-lactam are common tools in agriculture and can be found in animal-based foods such as goat milk and goat dried milk. To evaluate the risk of these species, reliable analytical methods are needed for accurate concentration determination, especially in goat milk and goat dried milk.MethodWe describe a method based on PRiME extraction coupled with UPLC-quadrupole/electrostatic field orbitrap high-resolution mass spectrometry to accomplish this task.ResultUnder optimal conditions, the limit of quantification for all antibiotics was 0.5–100 μg/L in goat milk and goat dried milk samples. The recoveries were 60.6–110.0% for goat milk and 60.1–109.6% for goat dried milk with a coefficient of variation less than 15%. The detection limits were 0.5–1.0 μg/kg. The limits of quantification for the analytes were 5.0–10.0 μg/kg. Finally, the method was used to screen veterinary antibiotics in 50 local goat milk and goat dried milk samples; metronidazole and enrofloxacin were detected in goat milk.ConclusionThis method offers good reliability and the capacity for simultaneous detection can be used to detect residual contents and evaluate health risks in goat milk and goat dried milk.

A 'Dilute and Shoot' Liquid Chromatography-Mass Spectrometry Method for Multiclass Drug Analysis in Pre-Cut Dried Blood Spots.

Drugs able to affect the auditory and nervous systems and consumed by workers to treatdifferent pathologies can represent a possible source of risk in the work environment. All the target compounds involved in the presented project show ototoxic and/or narcoleptic side effects and, for these reasons, occupational safety organizations have recognized them as potential causes of work injuries. A multiclass method for the analysis of 15 drugs among the most widespread worldwide (belonging to nine different classes including antihistamines, beta-blockers, antidepressants, Z-drugs and opioids), was developed and validated. This study describes a rapid, sensitive and effective method to analyse these substances in whole blood using tailored pre-cut dried blood spots. Detection was achieved with a triple quadrupole mass spectrometer after an easy and simple ‘dilute and shoot’ solubilisation followed by an UPLC separation. All the issues linked to the use of the dried blood spots and whole blood, such as haematocrit variability, volumetric evaluation and sample carrier choice were carefully studied and managed during method development. From the validation study results it emerged that this approach can be deemed successful thanks to its few pg µL−1 LOQs, good linear intervals, absolute recoveries of no less than 75%, an almost negligible matrix effect and accuracy and precision in line with the European and American guidelines for validation. All the obtained goals have been specifically pursued in order to encourage method diffusion as a primary prevention intervention, even in small private workplaces.

Non-linear Multiclass SVM Classification Optimization using Large Datasets of Geometric Motif Image

Support Vector Machine (SVM) with Radial Basis Functions (RBF) kernel is one of the methods frequently applied to nonlinear multiclass image classification. To overcome some constraints in the form of a large number of image datasets divided into nonlinear multiclass, there three stages of SVM-RBF classification process carried out i.e. 1) Determining the algorithms of feature extraction and feature value dimensions used, 2) Determining the appropriate kernel and parameter values, and 3) Using correct multiclass method for the training and testing processes. The OaO, OaA, and DAGSVM multi-class methods were tested on a large dataset of batik motif images whose geometric motifs with a variety of patterns and colors in each class and containing similar patterns in the motifs between the classes. DAGSVM has the advantage in classification accuracy value, i.e. 91%, but it takes longer during the training and testing processes.

Optimization of 204 veterinary drug residues method and establishing their mass spectrum library

ABSTRACT A multi-class multi-residue analytical method was validated in this paper for simultaneously identifying 204 veterinary drugs in pork by UPLC-QTRAP MS/MS, including sulfonamides, quinolones, β-agonists, and nitroimidazoles. MRM-IDA-EPI mode of QTRAP was used to establish a secondary mass spectrum library of these 204 veterinary drugs, which could enhance the qualitative analysis of suspect substances. The separation of substances was performed on a C18 column, with a gradient elution of 0.1% formic acid solution in water and methanol. Electrospray ionization (ESI), fast polarity switching, and scheduled MRM mode were used to collect the characterization results. Factors that affected the recovery were delicately optimized, such as solvent proportion, formic acid content, and purification method. We finally determined the extracting method with formic acid-acetonitrile-water (0.1:90:10, v:v:v) and purifying method with Cleanert LipoNo as a package for dispersive solid-phase extraction. Matrix effects on the quantitative results were also investigated. Some compounds had strong matrix inhibition or enhancement, and the matrix matching external standard method was used to quantify the compounds accurately. The correlation coefficient R2 was over 0.99 within the calibration linearity range for the detected 204 veterinary drugs. The limit of quantification (LOQ) was 0.5 ~ 5 μg/kg. The recoveries ranged from 61.4% to 118.8% with RSDs of 3.1% to 19.2% (n = 6).

Validation of a Multiclass Method for the Screening of 15 Antibiotic Residues in Milk Using Biochip Multi-array Technology and Its application to Monitor Real Samples.

Veterinary drugs are extensively and legally consumed to treat and prevent disease in chattels but some are also used illegally as growth-stimulating agents. Inappropriate or intensive use of antibiotics can cause allergic reactions and, above all, antibiotic resistance. A multiclass approach for the screening of antimicrobial substances in milk was validated in consonance with Commission Decision 2002/657/EC and to the European guideline for the validation of screening methods for veterinary medicines. This biochip-based approach enables the simultaneous determination of a total of 13 sulphonamide, dapsone and trimethoprim. For monitoring of antibiotic residues, 53 UHT milk samples collected from Tehran, IR Iran were screened applying this technology. The result showed that for all antibiotic residues, the positivity threshold T was much more than the cut-off value Fm. A false positive rate of less than 5% was found for all antibiotics which are satisfactory. All detection capabilities (CCβ) were well below the Maximum Residue Level (MRL) set by the European Commission (100 µg/kg for the sum of all sulphonamides and 50 µg/kg for trimethoprim in milk). The screening results of 53 milk samples showed that 71.7% of samples were compliant and all positive samples were below the MRL set by European Commission. This study showed that the biochip-based technique is valid to identify and quantify antibiotic residues in milk at the studied validation levels. The method was rapid, easy, safe, and able to screen 13 sulphonamide, dapsone and trimethoprim from a single milk sample simultaneously with no sample preparation procedure (or just one-step centrifugation).

Challenges and future directions in LC-MS-based multiclass method development for the quantification of food contaminants

Monitoring of food contaminants and residues has undergone a significant improvement in recent years and is now performed in an intensive manner. Achievements in the area of chromatography-mass spectrometry coupling techniques enabled the development of quantitative multi-target approaches covering several hundred analytes. Although the majority of methods are focusing on the analysis of one specific group of substances, such as pesticides, mycotoxins, or veterinary drugs, current trends are going towards the simultaneous determination of multiclass compounds from several families of contaminants and residues. This work provides an overview of relevant multiclass concepts based on LC-MS/MS and LC-HRMS instruments. Merits and shortcomings will be critically discussed based on current performance characteristics of the EU legislation system. In addition, the discussion of a recently developed multiclass approach covering >1000 substances is presented as a case study to illustrate the current developments in this area.

Development of a multiclass method to quantify phthalates, pharmaceuticals, and personal care products in river water using ultra-high performance liquid chromatography coupled with quadrupole hybrid Orbitrap mass spectrometry.

The organic micropollutants such as phthalates, pharmaceuticals, and personal care products (PPPCPs) enter the surface water through various routes. The aim of this study is to develop a sensitive and efficient method to identify and quantify 26 PPPCPs found in river water with acceptable accuracy and precision using a liquid chromatograph hyphenated with quadrupole hybrid Orbitrap mass spectrometry (Q-Orbitrap-MS) in a single chromatographic run. The organic micropollutants were extracted from river water by solid-phase extraction (SPE) using hydrophilic-lipophilic balance sorbent and analyzed using an ultra-high performance liquid chromatograph (UHPLC) equipped with C18 stationary phase for chromatographic separation. The targeted mass experiments were conducted in a Q-Orbitrap-MS system in positive and negative electrospray ionization mode. The method was found to be linear in the concentration range of 1-125ng/L with coefficient of determination lying in the range of 0.995-0.999. The method achieved limit of quantification in the range of 0.41-1.72ng/L, and method recovery measured at three different concentrations was found to be in the range of 75-115%. Intra- and interday precision expressed as percent relative standard deviation was found to be<15%. Matrix effect was found to be in the range of 83.5-109.79%. The matrix match calibration was used for quantification of PPPCPs in river water sample. The method performance was evaluated by analyzing real samples collected from Ganga River, and the concentrations of 21 analytes were found to be in the range of 0.76-9.49ng/L for pharmaceuticals, 1.49-8.67ng/L for phthalates, and 0.9-7.58ng/L for personal care products. The present method was found to be precise, sensitive, and rapid to determine 26 PPPCPs including phthalates in river water samples using SPE-UHPLC-Q-Orbitrap-MS.

Method optimization and validation for multi-class residue analysis in turmeric

Turmeric (Curcuma longa L.) is an economically important food and medicinal plant that grows in tropical and subtropical regions consumed worldwide. Turmeric is a difficult matrix for residue/contaminant analysis due to the high content of potentially interfering polyphenols (curcuminoids) and essential oils with physicochemical properties similar to target analytes. To solve this problem, a simple and straightforward sample preparation protocol based on ethyl acetate extraction and dispersive solid phase clean-up, followed by gas and liquid chromatography coupled to tandem mass spectrometry analysis, was optimized and validated, after selecting a representative blank material using gas chromatography coupled to ion mobility spectrometry and multivariate data analysis. Method validation was performed according to Codex guidelines for 67 pesticide residues, including 12 persistent organic pollutants, 4 aflatoxins and 3 dyes at 0.01, 0.02 and 0.05 mg kg−1level. Recoveries ranged between 60 and 120%, with relative standard deviations for repeatability and reproducibility below 20% and 32% respectively. Method performance was tested collaboratively between laboratories in Austria and Uruguay which helped proving the robustness of the methodology. The method was applied to the analysis of commercial samples and chlorpyrifos was detected in most of them.The novelty of the work relies in the optimization and validation of a multiclass method that allows analysis of four types of organic residues/contaminants, of paramount importance for food safety, using a single sample preparation step followed by gas and liquid chromatography coupled to tandem mass spectrometry determination.

Ensemble-support vector machine-random undersampling: Simulation study of multiclass classification for handling high dimensional and imbalanced data

Microarray technology measures on a large and parallel scale to express tens of thousands of genes. It has widely applied to predict gene function, new subtypes of specific tumors and cancer classification. However, microarray data are known has feature characteristics such as high dimension, small sample, high noise, and imbalanced class distribution. Support Vector Machine (SVM) has been widely used and shows the success in major applications to improve classification performance. To overcome the high dimension, we applied the Ensemble-SVM method. This method classifies features use clustering hierarchy and each group will be classified. While the condition of imbalance data becomes a problem in classification because the classifier will tend to predict the majority class compared to the minority class. Therefore, a Random Undersampling or EnSVM-RUS method is used to balance the size of the majority class into the minority class. We uses threefold cross-validation with a feature selection method that is Fast Correlation Based Filter (FCBF). The multiclass method used is SVM One Against One (OAO). While the evaluation criteria of performance classification based on the value of accuracy, F-score and G-mean and running time. We performs a simulation study with various scenario level of ratio imbalance (IR) that is ratio 1, 5, and 8 to know the performance of the proposed method. While the application on real data using Microarray DNA data with IR 4.22, 15.00 and 23.17 The results showed that EnSVM-RUS-OAO method with 2 clusters had higher performance than the EnSVM-OAO and EnSVM-OAO methods. Increasing the ratio imbalance doesn’t affect the advantage of the EnSVM-RUS-OAO method when compared to EnSVM-OAO and EnSVM-OAO methods. While on the use of the kernel, RBF kernel and polynomials produce higher performance and shorter computation time than linear kernels.

Investigation of the Correlation between the Use of Antibiotics in Aquaculture Systems and Their Detection in Aquatic Environments: A Case Study of the Nera River Aquafarms in Italy

Antimicrobial resistance (AMR) arises mostly because of the use and abuse of antibiotics in both the human and veterinary sectors. Furthermore, antibiotic residues can be discharged into the environment, promoting the spread of AMR as well as becoming a potential ecotoxicological risk. Aquaculture is one such source of environmental pollution. In the present work, we evaluated the correlation between antibiotic consumption in 11 selected aquafarms located along the Nera River, and the detection of their residues in surface water and sediment samples gathered monthly from the same stream. Antibiotic consumption was calculated using the ESVAC defined daily doses for animals (DDDvet) method, and a chemical analysis was conducted using a multiclass method. Therefore, the ecological risk of antibiotics being detected in surface waters was established based on the risk quotient (RQ). A strong positive correlation was identified between the concentrations detected in surface waters and the consumption of flumequine (r = 72%, p-value = 0.0085) and florfenicol (r = 83%, p-value = 0.0008). The RQ, however, was low for all the detected antibiotics in surface waters. Therefore, we proved that the antibiotics used in aquaculture can reach the near aquatic environment, but, in this work, they did not represent a toxicological risk to the surrounding ecosystem.

Analysis of antibiotic residues in raw bovine milk and their impact toward food safety and on milk starter cultures in cheese-making process

Antibiotics are widely used in livestock production as disease treatment, prevention and improve feeding efficiency. In this study three screenings and a multiclass method by Liquid Chromatography tandem Mass High Resolution Spectrometry (LC-HRMS) were developed and used to evaluate the antimicrobial substance incidence in 254 raw bovine milks involved in PDO cheese production in Northern Italy. The LC-HRMS results were compared to assess the frequency of false and negative results in term of reliability of screening methods. An investigation in relation to the quantified residues evaluating their possible negative impact on milk starter cultures in a simulated cheese-making process was presented. Lincomycin residues were observed in 30 samples, with 11.8% frequency and 17.29 ppb as mean value below its MRL. Three samples showed oxytetracycline respectively at 15.05, 0.82 and 1.59 ppb and two cefapirin and spyramycin at trace level. False positives with an acceptable frequency were observed by using the 3 kits confirming this approach useful for monitoring plans. Considering lincomycin, negative effect was demonstrated toward lactic acid bacteria activity in term of bacteria counts, pH and acidity during cheese making simulation. This represents a critical aspect considering both the economic value of PDO cheeses and the antibiotic resistance diffusion.

Off-line handwritten chinese character recognition by using support vector machines

The research proposes an efficient high-accuracy SVM-based handwritten Chinese character recognition system. The 304-dimensional feature vector of a handwritten Chinese character consists of the contour direction feature, the crossing count feature, the peripheral background area feature, and the contour line length feature. The mean-vector recognition method is first used as the coarse classifier to get a small number of candidate classes for the input vector. After the preliminary multi-class SVM-based recognition method is trained by 200 instances per candidate class, the recognition rate for the test handwritten characters of 5,401 classes can achieve 98.31%, which is much higher than about 93% accuracy for the mean-vector recognition method alone. To speed up the recognition, the forward and the backward greedy two-class SVM-based recognition methods are proposed to greatly reduce the recognition time to a practical level.

P-ODN: Prototype-based Open Deep Network for Open Set Recognition

Most of the existing recognition algorithms are proposed for closed set scenarios, where all categories are known beforehand. However, in practice, recognition is essentially an open set problem. There are categories we know called “knowns”, and there are more we do not know called “unknowns”. Enumerating all categories beforehand is never possible, consequently, it is infeasible to prepare sufficient training samples for those unknowns. Applying closed set recognition methods will naturally lead to unseen-category errors. To address this problem, we propose the prototype-based Open Deep Network (P-ODN) for open set recognition tasks. Specifically, we introduce prototype learning into open set recognition. Prototypes and prototype radiuses are trained jointly to guide a CNN network to derive more discriminative features. Then P-ODN detects the unknowns by applying a multi-class triplet thresholding method based on the distance metric between features and prototypes. Manual labeling the unknowns which are detected in the previous process as new categories. Predictors for new categories are added to the classification layer to “open” the deep neural networks to incorporate new categories dynamically. The weights of new predictors are initialized exquisitely by applying a distances based algorithm to transfer the learned knowledge. Consequently, this initialization method speeds up the fine-tuning process and reduce the samples needed to train new predictors. Extensive experiments show that P-ODN can effectively detect unknowns and needs only few samples with human intervention to recognize a new category. In the real world scenarios, our method achieves state-of-the-art performance on the UCF11, UCF50, UCF101 and HMDB51 datasets.