{"created":"2023-06-19T07:03:18.892659+00:00","id":5539,"links":{},"metadata":{"_buckets":{"deposit":"75d4d9db-78f3-4296-b6ab-8e71c9ad68bc"},"_deposit":{"created_by":3,"id":"5539","owners":[3],"pid":{"revision_id":0,"type":"depid","value":"5539"},"status":"published"},"_oai":{"id":"oai:rakuno.repo.nii.ac.jp:00005539","sets":["37:41:43"]},"author_link":["17448"],"item_3_biblio_info_9":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2018-06-30","bibliographicIssueDateType":"Issued"},"bibliographicPageEnd":"126","bibliographicPageStart":"1","bibliographic_titles":[{}]}]},"item_3_date_granted_67":{"attribute_name":"学位授与年月日","attribute_value_mlt":[{"subitem_dategranted":"2018-06-30"}]},"item_3_degree_grantor_65":{"attribute_name":"学位授与機関","attribute_value_mlt":[{"subitem_degreegrantor":[{"subitem_degreegrantor_name":"酪農学園大学"}],"subitem_degreegrantor_identifier":[{"subitem_degreegrantor_identifier_name":"30109","subitem_degreegrantor_identifier_scheme":"kakenhi"}]}]},"item_3_degree_name_64":{"attribute_name":"学位名","attribute_value_mlt":[{"subitem_degreename":"獣医学"}]},"item_3_description_43":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"subitem_description":"Thesis","subitem_description_type":"Other"}]},"item_3_description_6":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"と畜検査データとは、食肉衛生検査機関の獣医師であると畜検査員がと畜場に搬入された家畜を検査し、その結果をまとめたものである。と畜検査データは、衛生的な食肉生産促進のための生産者へのフィードバック事業や、調査研究に利用されてはいるものの、数値の羅列やグラフ表示を行う以上の処理が行われている例は非常に少ない。しかし、羅列された数値や折れ線グラフの傾きのみから飼養条件の悪化や改善の有無を客観的に判断することは困難である。すなわち、生産者、獣医師や行政官が合理的に意思決定するための指標を与える統計学的手法が必要とされていた。と畜検査データは時間とともに出現する時系列データであり、時系列分析に適していると考えられる。しかし、時系列分析には種々の方法があり、データの分布に応じた適切なモデルを用いる必要がある。そこで本研究は、データの統計学的分布に応じてモデルを使い分け、網羅的に分析できる手法を見出すことを目的とした。第一章では、常時発生がみられる疾病の分析方法として、代表的なガウス型時系列モデルである季節自己回帰和分移動平均モデル(SARIMAモデル)によると畜検査データの分析方法と、これを利用して農場における衛生状態や対策の効果を評価する方法について述べた。第二章では、発生が稀な疾病の時系列分析を行う方法として、周期的要素について擬似ポアソン回帰モデルを推定し、その残差についてARIMAモデルを推定したのち、両モデルを加える方法について述べた。第三章では、発生が稀な疾病の時系列分析を行うもう一つの方法として、two-partモデルを用いた分析方法について検討した。第一章では、SARIMAモデルを用いてと畜検査における廃棄率データの異常な増加や減少を検出できる好適条件を検討した． 寄生虫性肝炎による肝臓廃棄率が30%以上に増加した後、駆虫剤の投与により1.8%まで低下した1農場のデータを用い、廃棄率の増加および減少を検出できる最大の信頼水準を調べたところ、85%が至適条件であった。さらに、豚で常時発生がみられる4疾病について10農場からの搬入豚の検査データを用い， SARIMAモデルの当てはまりと対照モデルである指数平滑法の当てはまりを比較した． 40モデル中39モデルで対照モデルより逸脱度が小さく， 27モデルで有意であった． 本法は生産者に対して農場の衛生状態の合理的な判断指標を提供することによって， 衛生的な食肉生産を促進するツールとして利用可能であることが明らかとなった。第二章では、人獣共通感染症である豚の非定型抗酸菌症による毎日の肝臓廃棄数データを時系列分析し、農場アウトブレイクの疑いがある生産者からの搬入の有無を検出する方法について検討した。豚の非定型抗酸菌症は、比較的稀な疾病であり、廃棄数がゼロであることも少なくないため、第一章の分析方法は適さない。そこで、擬似ポアソン回帰モデルとARIMAモデルを組み合わせる方法を試みた。10年間の日次廃棄数値のうち最初の8年分のデータを用いてモデルを作成し、残りの2年分の日次廃棄数がモデルによる期待値の95%信頼限界を超えた日の搬入生産者を調べたところ、当該搬入日には年次廃棄数集計値の廃棄率から見て抗酸菌症の農場アウトブレイクが疑われる生産者による搬入が必ず行われていた。すなわち、本モデルにより発生が稀な疾病の時系列分析を行い、農場アウトブレイクを検出することが可能となった。第三章では、人獣共通感染症である豚のエキノコックス症による月次肝臓廃棄数データを時系列分析し、農場アウトブレイクを検出する方法について検討した。北海道東藻琴食肉衛生検査所では、月次集計データにおける廃棄率が1%を超えるか、3か月連続でエキノコックス症による廃棄が認められた場合に、対策を促す情報提供を行っている。しかし、アウトブレイクの定義を考慮し、判定は統計学的に算出された期待値との比較によって行われるべきである。ここで問題となるのは、エキノコックス症は比較的稀な疾病であるために、ゼロデータが多く、毎月のルーチン業務で用いるには二章の方法のような煩雑な方法は適さないことである。そこで、two-partモデルによって計算されたパーセンタイル値をアウトブレイク検出の基準とすることを検討したところ、従来の基準よりも迅速なアウトブレイク検出が可能であることが明らかとなった。すなわち、本モデルにより発生が稀な疾病の時系列分析をワンステップで行い、アウトブレイクを検出することが可能であることが明らかとなった。以上をまとめる。廃棄率が極端に0%や100%付近に偏っていなければ、ロジット変換した廃棄率は、正規分布することを仮定できるため、第一章に述べたSARIMAモデルを用いた分析を行うのが適切と考えられる。ゼロの多いデータについては、第二章の方法と第三章の方法を用いることができるが、ルーチンの検査業務で用いることを考慮すると、ワンステップでモデリングが可能なtwo-partモデルの方が更新を容易に行うことができ、実用的であろう。すなわち、廃棄数がゼロであることが稀な疾病については、第一章のSARIMAモデルで分析し、ゼロが多い場合はtwo-partモデルで分析することで、いかなる疾病に対しても客観的指標に基づく情報提供を可能にする道筋を付けることが出来た。","subitem_description_type":"Abstract"},{"subitem_description":"Meat inspection data summarizes and integrates the results of meat inspections at slaughterhouses conducted by official veterinarians with local government meat inspection centers. Although meat inspection data has been utilized for surveys and research and fed back to producers to improve hygiene in meat production, few cases have been reported in which anything more than listing or drawing graphs of meat inspection data is performed. However, it is difficult to determine objectively whether livestock breeding conditions tend to become better or worse based on the enumeration of figures or quantity of change in graphs. That is, statistical methods that provide criteria on which the decisions made by producers, veterinarians, and administrators are based are needed. Time series analysis is a suitable method to analyze meat inspection data as it comprises sequential time series data. Since time series analysis involves a variety of methods, to analyze meat inspection data, it is necessary to select a proper method based on the statistical distribution of data. Therefore, the purposes of the present study were to identify the optimal method for using multiple models properly according to the distribution of data and to analyze data with different types of distributions comprehensively. Chapter I describes the time series analysis method for meat inspection data using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model, which is a typical Gaussian time series model for the analysis of diseases regularly seen that describes methods for estimating the sanitation or effectiveness of hygienic measures at farms. Chapter II describes the method for analyzing diseases rarely seen by estimating, and subsequently merging, a quasi-Poisson regression model for cyclic components and the ARIMA (Autoregressive Integrated Moving Average) model for residuals. Chapter III describes another method for analyzing diseases rarely seen using the two-part model. In Chapter I, investigations are carried out to obtain the favorable conditions for the SARIMA model to detect anomalies in the condemnation rate. The data of one farm, in which the condemnation rate of liver due to parasitic hepatitis was elevated above 30% and reduced to 1.8% by anthelminthic administration, was used to determine the maximum confidence level value able to detect the elevation and reduction of the condemnation rate. As a result, the optimum confidence level value was 85%. In addition, the degree of fit between the condemnation rates and the models estimated using the SARIMA model and those estimated using the exponential smoothing method as a control were compared for four diseases regularly seen in swine from 10 producers. The deviance for 39 of 40 models was smaller than the control, and statistical significance was shown for 27 of 40 models. It became clear that this method can provide an index for determining the sanitation of farms for producers and be utilized as a tool to promote hygienic meat production. In Chapter II, a time series analysis of the daily number of liver condemnations due to atypical mycobacterial disease in swine was conducted and the optimal method to determine whether swine were brought to the slaughterhouse on a specific day from the farm at which the mycobacteriosis outbreak was suspected was investigated. Because atypical mycobacteriosis in swine is relatively rare and the number of condemnations due to atypical mycobacteriosis is often zero, the method described in Chapter I was not suitable for a time series analysis of the disease. Accordingly, the method to merge the quasi-Poisson regression and ARIMA models was tested. The model was developed using daily inspection data from the first 8 of 10 years. Then it was investigated which producers brought pigs to slaughterhouses when the daily number of condemned livers exceeded the 95% confidence interval of the expected values calculated by the model during the remaining two years Subsequently, on the day on which the confidence interval was exceeded, the producers bringing pigs to the slaughterhouses always included the producers at farms at which farmlevel outbreaks were highly suspected because of the high annual rates of condemnation due to mycobacteriosis. The above results revealed that the time series model enabled the detection of farm-level outbreaks using a time series analysis of rarely seen diseases. In Chapter III, the method to detect farm-level outbreaks of swine echinococcosis, which is one of the most lethal zoonoses, using a time series analysis of the number of monthly condemned livers due to echinococcosis was investigated. Hokkaido Higashi-Mokoto Meat Inspection Center provides farms with a brochure detailing preventive measures to take when the condemnation of livers due to echinococcosis exceeds 1% of carcasses or when more than one pig carcass is condemned for three consecutive months. However, considering the definition of an outbreak, the determination must be made based on a comparison between the observed and expected values calculated using statistical methods. The problems are that the monthly echinococcosis data for each farm contains many zeros because swine echinococcosis is a rare disease, and that the method described in Chapter II is too complex to perform routine modeling every month. To solve these problems, the method for utilizing the percentiles calculated by the time series model based on a two-part model as criteria to detect outbreaks was investigated. As a result, the time series model was able to detect outbreaks earlier than the conventional criteria. Therefore, the model described could perform a time series analysis of a rare disease in a single step and detect farm-level outbreaks. In conclusion, the SARIMA model described in Chapter I is suitable for analyzing condemnation rate data that do not deviate excessively around 0% or 100%, as the logittransformed condemnation rate is assumed to be normally distributed. These results suggest that data containing many zeros can be analyzed using the methods described Chapters II and III. To analyze data in routine meat inspections, the two-part model described in Chapter III is more suitable because it is easier to renew models that are estimated in a single step. Therefore, the present study paves the way to enable providing administrative information regarding various diseases based on objective criteria using the SARIMA model for diseases in which the data rarely contain zeros and the two-part model for diseases in which the data contain many zeros.","subitem_description_type":"Abstract"}]},"item_3_dissertation_number_68":{"attribute_name":"学位授与番号","attribute_value_mlt":[{"subitem_dissertationnumber":"乙第141号"}]},"item_3_version_type_19":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_version_resource":"http://purl.org/coar/version/c_970fb48d4fbd8a85","subitem_version_type":"VoR"}]},"item_access_right":{"attribute_name":"アクセス権","attribute_value_mlt":[{"subitem_access_right":"open access","subitem_access_right_uri":"http://purl.org/coar/access_right/c_abf2"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"足立, 泰基"}],"nameIdentifiers":[{}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2018-08-20"}],"displaytype":"detail","filename":"adachi_hakuron.pdf","filesize":[{"value":"2.9 MB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"博士論文本文","url":"https://rakuno.repo.nii.ac.jp/record/5539/files/adachi_hakuron.pdf"},"version_id":"c4aa3f83-1575-41b4-bb97-bf604026e316"},{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2018-08-20"}],"displaytype":"detail","filename":"adachi_youshi.pdf","filesize":[{"value":"171.4 kB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"論文要旨と審査結果の要旨","url":"https://rakuno.repo.nii.ac.jp/record/5539/files/adachi_youshi.pdf"},"version_id":"ff7ebd3c-7629-42bf-9a01-bd58443cfd75"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"doctoral thesis","resourceuri":"http://purl.org/coar/resource_type/c_db06"}]},"item_title":"と畜検査データの時系列分析に関する研究","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"と畜検査データの時系列分析に関する研究"}]},"item_type_id":"3","owner":"3","path":["43"],"pubdate":{"attribute_name":"公開日","attribute_value":"2018-08-20"},"publish_date":"2018-08-20","publish_status":"0","recid":"5539","relation_version_is_last":true,"title":["と畜検査データの時系列分析に関する研究"],"weko_creator_id":"3","weko_shared_id":-1},"updated":"2023-06-19T08:01:05.992648+00:00"}