Towards Application of One-Class Classification Methods to Medical Data

In the problem of one-class classification (OCC) one of the classes, the target class, has to be distinguished from all other possible objects, considered as nontargets. In many biomedical problems this situation arises, for example, in diagnosis, image based tumor recognition or analysis of electrocardiogram data. In this paper an approach to OCC based on a typicality test is experimentally compared with reference state-of-the-art OCC techniques-Gaussian, mixture of Gaussians, naive Parzen, Parzen, and support vector data description-using biomedical data sets. We evaluate the ability of the procedures using twelve experimental data sets with not necessarily continuous data. As there are few benchmark data sets for one-class classification, all data sets considered in the evaluation have multiple classes. Each class in turn is considered as the target class and the units in the other classes are considered as new units to be classified. The results of the comparison show the good performance of the typicality approach, which is available for high dimensional data; it is worth mentioning that it can be used for any kind of data (continuous, discrete, or nominal), whereas state-of-the-art approaches application is not straightforward when nominal variables are present.

Comparison of survey methods for estimating abundance of harbor seals (Phoca vitulina) in glacial fjords

The importance of glacial ice habitats to harbor seals (Phoca vitulina) in Alaska has become increasingly apparent. However, enumerating harbor seals hauled out on ice in glacial fjords has been difficult. At Johns Hopkins Inlet in Glacier Bay, Alaska, we compared a shore-based counting method to a large-format aerial photography method to estimate seal abundance. During each aerial survey, shore-based observers simultaneously counted seals from an observation post. Both survey methods incurred errors in double-counting and missing seals, especially when ice movements caused seals to drift between survey zones. Advantages of shore-based counts included the ability to obtain multiple counts for relatively little cost, distinguish pups from adults, and to distinguish mobile seals from shadows or glacial debris of similar size. Aerial photography provided a permanent record of each survey, allowing both a reconciliation of counts in overlapping zones and the documentation of the spatial distribution of seals and ice within the fjord.

Evaluating methods for estimating rare events with zero-heavy data: a simulation model estimating sea turtle bycatch in the pelagic longline fishery

Estimating rare events from zero-heavy data (data with many zero values) is a common challenge in fisheries science and ecology. For example, loggerhead sea turtles (Caretta caretta) and leatherback sea turtles (Dermochelys coriacea) account for less than 1% of total catch in the U.S. Atlantic pelagic longline fishery. Nevertheless, the Southeast Fisheries Science Center (SEFSC) of the National Marine Fisheries Service (NMFS) is charged with assessing the effect of this fishery on these federally protected species. Annual estimates of loggerhead and leatherback bycatch in a fishery can affect fishery management and species conservation decisions. However, current estimates have wide confidence intervals, and their accuracy is unknown. We evaluate 3 estimation methods, each at 2 spatiotemporal scales, in simulations of 5 spatial scenarios representing incidental capture of sea turtles by the U.S. Atlantic pelagic longline fishery. The delta-log normal method of estimating bycatch for calendar quarter and fishing area strata was the least biased estimation method in the spatial scenarios believed to be most realistic. This result supports the current estimation procedure used by the SEFSC.

Application of two methods for determining diet of northern fur seals (Callorhinus ursinus)

Examination of hard parts recovered from scats (feces) is currently the most common method for determining the diet of pinnipeds. However, large or sharp prey remains may be spewed (regurgitated) biasing prey composition and size estimations in diet studies based on scats. Percent frequency of occurrence (FO%) and age or size of selected prey remains recovered from northern fur seal (Callorhinus ursinus) scat (n=3444) and spew samples (n=267) collected from rookeries on St. George Island and St. Paul Island, Alaska, between 1990 and 2000 were compared to determine if a bias in prey composition and age or size estimations existed between scats and spews. Overall prey composition was similar between sample type and location, but the relative FO% of primary prey (≥5%) varied by sample type and location. Age or size estimates of walleye pollock (Theragra chalcogramma) and of two species of gonatid squids (Gonatopsis borealis and Berryteuthis magister) were significantly larger in spews than in scats. Observed differences in FO% and estimated age or size of prey species whose remains were found in scats and spews likely result from size-selective digestion of prey remains. Scats were biased toward smaller prey remains, whereas spews were biased toward larger prey remains and cephalopod beaks. The percent overlap between age classes of walleye pollock caught by the commercial trawl fishery and age classes of walleye pollock consumed by northern fur seals varied noticeably between sample types for both islands (scats: St. George=15. 5%; St. Paul=4.1%; spews: St. George=94.6%; St. Paul=89.6%). These results demonstrate that the inclusion of multiple sampling methods allows for a more accurate assessment of northern fur seal prey occurrence and prey age and size.

Growth and maturity of salmon sharks (Lamna ditropis) in the eastern and western North Pacific, and comments on back-calculation methods

Age and growth estimates for salmon sharks (Lamna ditropis) in the eastern North Pacific were derived from 182 vertebral centra collected from sharks ranging in length from 62.2 to 213.4 cm pre-caudal length (PCL) and compared to previously published age and growth data for salmon sharks in the western North Pacific. Eastern North Pacific female and male salmon sharks were aged up to 20 and 17 years, respectively. Relative marginal increment (RMI) analysis showed that postnatal rings form annually between January and March. Von Bertalanffy growth parameters derived from vertebral length-at-age data are L∞ =207.4 cm PCL, k=0.17/yr, and t0=−2.3 years for females (n=166), and L∞ =182.8 cm PCL, k=0.23/yr , and t0=−1.9 years for males (n=16). Age at maturity was estimated to range from six to nine years for females (median pre-caudal length of 164.7 cm PCL) and from three to five years old for males (median precaudal length of 124.0 cm PCL). Weight-length relationships for females and males in the eastern North Pacific are W=8.2 × 10_05 × L2.759 –06 × L3.383 (r2 =0.99) and W=3.2 × 10 (r2 =0.99), respectively. Our results show that female and male salmon sharks in the eastern North Pacific possess a faster growth rate, reach sexual maturity earlier, and attain greater weight-at-length than their same-sex counterparts living in the western North Pacific.

A study of tagging methods for the sea cucumber Cucumaria frondosa in the waters off Maine

The sea cucumber fishery in waters off Maine is developing and has recently experienced great increases in landings, corresponding to expanding export markets. Between 1994 and 1996, reported landings ranged from one to three million pounds (Fig. 1). In 1999, reported landings were over eight million pounds and rose to over nine million in 2000 (Feindel1). Like other developing fisheries, we have little information about the biology and ecology of the sea cucumber off Maine, limited data on the fishery, and little knowledge about the key life history processes that characterize its population dynamics. Therefore, we have a limited understanding of the current status of the resource and the impacts the fishery may have on the stock.

Precision and accuracy of fish length measurements obtained with two visual underwater methods

During the VITAL cruise in the Bay of Biscay in summer 2002, two devices for measuring the length of swimming fish were tested: 1) a mechanical crown that emitted a pair of parallel laser beams and that was mounted on the main camera and 2) an underwater auto-focus video camera. The precision and accuracy of these devices were compared and the various sources of measurement errors were estimated by repeatedly measuring fixed and mobile objects and live fish. It was found that fish mobility is the main source of error for these devices because they require that the objects to be measured are perpendicular to the field of vision. The best performance was obtained with the laser method where a video-replay of laser spots (projected on fish bodies) carrying real-time size information was used. The auto-focus system performed poorly because of a delay in obtaining focus and because of some technical problems.

Biochemical biomarkers in samples processed by different methods in Environmental Specimen Banks

In the last decades the creation of new Environmental Specimen Banks (ESB) is increasing due to the necessity of knowing the effects of pollutants in both the environment and human populations. ESBs analyze and store samples in order to understand the effects of chemicals, emerging substances and the environmental changes in biota. For a correct analysis of the effect induced by these variables, there is a need to add biological endpoints, such as biomarkers, to the endpoints based on chemical approaches which have being used until now. It is essential to adapt ESB´s sampling strategies in order to enable scientists to apply new biological methods. The present study was performed to obtain biochemical endpoints from samples stored in the BBEBB (Biscay Bay Environmental Biospecimen Bank) of the Marine Station of Plentzia (PIE - UPV/EHU). The main objective of the present work was to study the variability caused in biochemical biomarkers by different processing methods in mussels (Mytilus galloprovincialis) from two localities (Plentzia and Arriluze) with different pollution history. It can be concluded that the selected biomarkers (glutathione S-transferase and acetylcholinesterase) can be accurately measured in samples stored for years in the ESBs. The results also allowed the discrimination of both sampling sites. However, in a further step, the threshold levels and baseline values should be characterized for a correct interpretation of the results in relation to the assessment of the ecosystem health status.

A simple spreadsheet model to incorporate seasonal growth into length-based stock assessment methods

The paper describes a method by which seasonal growth can be incorporated into length-converted catch curves and cohort analyses using a spreadsheet. The method is based on calculating the length of fish using seasonal growth parameters on a daily basis. A LOOKUP function is then used to find the age corresponding to the length.

Monte Carlo methods for TMD analyses

4th International Workshop on Transverse Polisarization Phenomena in Hard Processes (TRANSVERSITY 2014)