Search

Despite the advancement of machine learning techniques in recent years, state-of-the-art systems lack robustness to "real world" events, where the input distributions and tasks encountered by the deployed systems will not be limited to the original training context, and systems will instead need to adapt to novel distributions and tasks while deployed. This critical gap may be addressed through the development of "Lifelong Learning" systems that are capable of (1) Continuous Learning, (2) Transfer and Adaptation, and (3) Scalability. Unfortunately, efforts to improve these capabilities are typically treated as distinct areas of research that are assessed independently, without regard to the impact of each separate capability on other aspects of the system. We instead propose a holistic approach, using a suite of metrics and an evaluation framework to assess Lifelong Learning in a principled way that is agnostic to specific domains or system techniques. Through five case studies, we show that this suite of metrics can inform the development of varied and complex Lifelong Learning systems. We highlight how the proposed suite of metrics quantifies performance trade-offs present during Lifelong Learning system development - both the widely discussed Stability-Plasticity dilemma and the newly proposed relationship between Sample Efficient and Robust Learning. Further, we make recommendations for the formulation and use of metrics to guide the continuing development of Lifelong Learning systems and assess their progress in the future., Competing Interests: Declaration of Competing Interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Aswin Nadamuni Raghavan has patent #US16/825,953 pending to SRI International Inc. Mario Aguilar-Simon, Andrew Brna, Ryan Brown has patent #US20200334579A1 pending to Teledyne Scientific & Imaging, LLC. Nicholas Ketz, Neale Ratzlaf, and Praveen Pilly have patents pending to HRL Laboratories, LLC., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Ancylostoma caninum is the most prevalent intestinal nematode of dogs, and has a zoonotic potential. Multiple-drug resistance (MDR) has been confirmed in a number of A. caninum isolates, including isolate Worthy 4.1F3P, against all anthelmintic drug classes approved for hookworm treatment in dogs in the United States (US). The cyclooctadepsipeptide emodepside is not registered to use in dogs in the US, but in a number of other countries/regions. The objective of this study was to evaluate the efficacy of emodepside + praziquantel, as well as three commercial products that are commonly used in the US for treatment of hookworms, against a suspected (subsequently confirmed) MDR A. caninum isolate Worthy 4.1F3P. 40 dogs infected on study day (SD) 0 with 300 third-stage larvae, were randomly allocated to one of five treatment groups with eight dogs each: pyrantel pamoate (Nemex®-2), fenbendazole (Panacur® C), milbemycin oxime (Interceptor®), emodepside + praziquantel tablets and non-treated control. Fecal egg counts (FEC) were performed on SDs 19, 20, 22, 27, 31 and 34. All treatments were administered as per label requirements on SD 24 to dogs in Groups 1 through 4. Two additional treatments were administered on SDs 25 and 26 to dogs in Group 2 as per label requirements. Dogs were necropsied on SD 34 and the digestive tract was removed/processed for worm recovery and enumeration. The geometric mean (GM) worm counts for the control group was 97.4, and for the pyrantel pamoate, fenbendazole, milbemycin oxime, and emodepside + praziquantel groups were 74.8, 72.0, 88.9, and 0.4, respectively. These yielded efficacies of 23.2%, 26.1%, and 8.8%, and 99.6%, respectively. These data support previous findings of the MDR status of Worthy 4.1F3P as treatments with pyrantel pamoate, fenbendazole and milbemycin oxime lacked efficacy. In sharp contrast, Worthy 4.1F3P was highly susceptible to treatment with emodepside + praziquantel., Competing Interests: Declaration of competing interest The study reported here was funded by Bayer US LLC, Animal Health, Research and Development, Shawnee, KS, USA and was performed at TRS Labs Inc., (Copyright © 2020 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Plasmodium vivax is now the predominant cause of malaria in the Asia-Pacific, South America and Horn of Africa. Laboratory studies of this species are constrained by the inability to maintain the parasite in continuous ex vivo culture, but genomic approaches provide an alternative and complementary avenue to investigate the parasite's biology and epidemiology. To date, molecular studies of P. vivax have relied on the Salvador-I reference genome sequence, derived from a monkey-adapted strain from South America. However, the Salvador-I reference remains highly fragmented with over 2500 unassembled scaffolds.  Using high-depth Illumina sequence data, we assembled and annotated a new reference sequence, PvP01, sourced directly from a patient from Papua Indonesia. Draft assemblies of isolates from China (PvC01) and Thailand (PvT01) were also prepared for comparative purposes. The quality of the PvP01 assembly is improved greatly over Salvador-I, with fragmentation reduced to 226 scaffolds. Detailed manual curation has ensured highly comprehensive annotation, with functions attributed to 58% core genes in PvP01 versus 38% in Salvador-I. The assemblies of PvP01, PvC01 and PvT01 are larger than that of Salvador-I (28-30 versus 27 Mb), owing to improved assembly of the subtelomeres.  An extensive repertoire of over 1200 Plasmodium interspersed repeat ( pir ) genes were identified in PvP01 compared to 346 in Salvador-I, suggesting a vital role in parasite survival or development. The manually curated PvP01 reference and PvC01 and PvT01 draft assemblies are important new resources to study vivax malaria. PvP01 is maintained at GeneDB and ongoing curation will ensure continual improvements in assembly and annotation quality., Competing Interests: No competing interests were disclosed.

Background: This controlled laboratory study was designed to evaluate the efficacy of the 10 % imidacloprid/4.5 % flumethrin collar (Seresto®, Bayer Animal Health) against fleas (Ctenocephalides f. felis) on cats, when compared to fipronil (9.8 %w/w)/(s)-methoprene (11.8 % w/w) topical spot-on formulation (Frontline® Plus for Cats and Kittens, Merial)., Methods: Thirty cats were randomized into three groups of ten animals based on pre-treatment flea counts: Group 1: imidacloprid/flumethrin collar; Group 2: fipronil/(s)-methoprene topical spot-on and Group 3: non-treated controls. The imidacloprid/flumethrin collars were applied one time on Day 0, while the fipronil/(s)-methoprene spot-on was administered every 30 days from Day 0 through Day 210. Cats were infested with 100 fleas on study days 0, 7, 14, 29, 59, 89, 119, 149, 179, 209 and 239. All flea counts were conducted by combing to remove fleas on post-treatment days 2, 8, 15, 30, 60, 90, 120, 150, 180, 210 and 240., Results: The efficacy of the imidacloprid/flumethrin collar ranged from 98.2 to 100 % for eight months. The efficacy of fipronil/(s)-methoprene spot-on ranged from 68.2 to 99.9 %. Efficacy was < 85 % for fipronil/(s)-methoprene on Days 90, 150 and 210. The flea counts in both treatment groups were significantly fewer than those in the non-treated control group at every post-treatment study day (P < 0.0001). In addition, there were significantly fewer fleas in the imidacloprid/flumethrin collar group when compared to the fipronil/(s)-methoprene group on Days 90, 150 and 210 (P < 0.0001)., Conclusions: This study demonstrated that the imidacloprid/flumethrin collar (Seresto®, Bayer Animal Health) maintained excellent ( > 98.2 %) efficacy against fleas on cats for the entire 8 month study. Monthly applications of fipronil/(s)-methoprene (Frontline® Plus for Cats and Kittens, Merial) generally had high, but variable (68.2 to 99.9 %) efficacy over the course of the eight month study. Based on the very high residual efficacy achieved by the imidacloprid/flumethrin collar in this study, veterinarians should expect that this collar will control and eliminate existing flea infestations on cats and in their in-home premises as long as every flea infested host is treated.

This controlled laboratory study demonstrated the residual speed of efficacy of an imidacloprid/flumethrin collar (Seresto(®), Bayer) for the control of ticks (Dermacentor variabilis, Amblyomma americanum) at 6 and 12 hours post-infestation on dogs when compared to oral fluralaner (Bravecto(®), Merck). Dogs were randomised by pre-treatment tick counts: Group 1) imidacloprid 10 % (w/w)/flumethrin 4.5 % (w/w) collar, 2) fluralaner (dosage 25.1 - 49.4 mg/kg), and 3) non-treated controls. Ticks (50/species/dog) were infested on days 3, 14, 21, 28, 42, and 56 followed by 50 D. variabilis on days 70 and 84. Live and dead attached ticks were counted 6 and 12 hours later. Efficacy against both species at 6 and 12 hours for Group 1 was 94 - 100 %. Efficacy for Group 2 against both species at 6 hours was 4 - 69 %; efficacy at 12 hours was 8 - 100 %. Live (attached and non-attached) tick counts at 6 hours in Group 1 were significantly lower (p ≤ 0.05) than counts in Group 2 and 3 on all days. At 12 hours, live counts were significantly lower (p ≤ 0.05) in Group 1 than Group 2 for D. variabilis from days 56 - 84 and for A. americanum from days 28 - 56. There were significantly fewer (p ≤ 0.05) total ticks (total live and dead attached) on dogs in Group 1 compared to Group 2 and 3 at all time points. This study demonstrated that an imidacloprid/flumethrin collar was highly efficacious (94 - 100 %) at repelling and killing ticks on dogs at 6 and 12 hours post-infestation and was more efficacious than fluralaner as early as 6 hours post-infestation on all challenge days.

The susceptibility of 12 field-collected isolates and 4 laboratory strains of cat fleas, Ctenocephalides felis was determined by topical application of some of the insecticides used as on-animal therapies to control them. In the tested field-collected flea isolates the LD50 values for fipronil and imidacloprid ranged from 0.09 to 0.35 ng/flea and 0.02 to 0.19 ng/flea, respectively, and were consistent with baseline figures published previously. The extent of variation in response to four pyrethroid insecticides differed between compounds with the LD50 values for deltamethrin ranging from 2.3 to 28.2 ng/flea, etofenprox ranging from 26.7 to 86.7 ng/flea, permethrin ranging from 17.5 to 85.6 ng/flea, and d-phenothrin ranging from 14.5 to 130 ng/flea. A comparison with earlier data for permethrin and deltamethrin implied a level of pyrethroid resistance in all isolates and strains. LD50 values for tetrachlorvinphos ranged from 20.0 to 420.0 ng/flea. The rdl mutation (conferring target-site resistance to cyclodiene insecticides) was present in most field-collected and laboratory strains, but had no discernible effect on responses to fipronil, which acts on the same receptor protein as cyclodienes. The kdr and skdr mutations conferring target-site resistance to pyrethroids but segregated in opposition to one another, precluding the formation of genotypes homozygous for both mutations.

This randomised controlled laboratory study demonstrated the residual speed of efficacy of an imidacloprid/flumethrin collar (Seresto(®), Bayer) for the control of ticks (Dermacentor variabilis, Amblyomma americanum) at 6 and 12 hours postinfestation on dogs when compared to oral afoxolaner (NexGard(®), Merial). Dogs were randomised by pre-treatment tick counts: Group 1) imidacloprid 10 % (w/w) / flumethrin 4.5 % (w/w) collar, 2) afoxolaner chewable (dosage 3.1 - 6.2 mg/kg), and 3) non-treated controls. Ticks (50/species/dog) were infested on days 3, 14, 21, and 28; live (attached and non-attached) and dead attached ticks were counted 6 and 12 hours later. Efficacy against live D. variabilis at 6 hours for Group 1 was 95 - 100 % and for Group 2 was 38 - 48 %; efficacy at 12 hours for Group 1 was 97 - 100 % and for Group 2 was 27 - 59 %. Efficacy against A. americanum at 6 hours for Group 1 was 94 - 100 % and for Group 2 was < 0 - 38 %; efficacy at 12 hours for Group 1 was 98 - 100 % and for Group 2 was 1 - 40 %. Live and total (total live and dead attached) tick counts in Group 1 against both tick species were significantly lower (p ≤ 0.05) than Group 2 and 3 at all time points. The number of live or total ticks on Group 2 dogs was never significantly lower when compared to the respective number of ticks on Group 3 (controls). This study demonstrated that an imidacloprid/flumethrin collar was highly efficacious (94 - 100 %) at repelling and killing ticks on dogs at 6 and 12 hours post-infestation and was more efficacious than afoxolaner on all challenge days.