当社のテスト結果を提示します。時間が許せば、国内学会で行なった、このワーキンググループに関連する発表について少しだけご紹介できればと思っています。まず、テスト結果と各用語の分類（Scope）の提案について説明します。以前ご説明いたしましたとおり、我々は、A、B、Cの3つをpositive controlに、DとEをnegative controlとしてテストいたしました。863個ある用語はFとGの2つのグループに分けました。こちらが結果です。まず、グループFです。
評価
52
翻訳 / 英語
- 2016/08/31 15:48:50に投稿されました
I will show our test result. If time allows, I would like to introduce a little about the presentation related to this working group, which was held at a domestic academic conference. Firstly, let me explain on the test result and a proposal to the classification (scope) of each term. As explained before, we have examined separating three A, B, C into positive control, and D and E into negative control. We separated 863 terms into two groups, F and G. Here is the result. Firstly, this is group F.
評価
50
翻訳 / 英語
- 2016/08/31 16:12:33に投稿されました
We show test result of our company. If we have time, we will introduce only a part of presentation in regard with this working group that was held in academic meeting in Japan. We will explain result of the test and suggestion of classification (scope) of each term. As we had explained before, we carried out the test by classifying A,B and C as positive control as well as D and D as negative control. 863 terms were classified into F and G. This is a test result. First is group F.
結果を色分けしてありますが、一番濃い10個の用語は、PRR3以上かつ３つのpositive controlのcompound全てで報告数が3件以上あるもの、次に濃いものは2つのcompound、その次が1つのcompoundで、という順番です。一番薄いtermは、これが一番多いのですが、PRRが２未満か、いずれのcompoundでも報告件数が2件以下のものです。”Scope”の欄に、当社の提案を入れてあります。
![[削除済みユーザ]](https://secure.gravatar.com/avatar/695023b4efd563721d13716b1457bc73.png?d=identicon&mode=crop&r=PG&s=42)
評価
51
翻訳 / 英語
- 2016/08/31 15:49:53に投稿されました
The results are divided into each colors but 10 strong colors of the words are that the number of the reoorts is more than 3, and more than 3 PRR. And the next strong colors have 2 compounds and the next colors have 1 compound.
And the weakest colors are less than 2 PER or less than 2 reports.
Our suggestions are in the "Scooe" box.
And the weakest colors are less than 2 PER or less than 2 reports.
Our suggestions are in the "Scooe" box.
評価
50
翻訳 / 英語
- 2016/08/31 16:06:55に投稿されました
The result is classified by color. The strongest 10 terms are the one where there are more than 3 PRRs as well as number of report is more than 3 by all compounds of 3 positive controls. Next strongest is 2 compounds, next after that is 1 compound. The lightest term is the greatest in number. It is the one where either there are fewer than 2 PRRs or number of report by any compound is fewer than 2. In the column of "scope", we input idea of our company.
第一段階として、次のようにルールを決め、機械的にScopeを当てはめました。グループFでは、PRRが2以上かつ、いずれかのcompoundで報告数が3以上の用語を＃＃としました。
残りの一番薄い色の部分ですが、Tomasの作業と重複しますが、まず検査名のみの用語や陰性、偽陽性が明示されている用語は＊＊としました。それ以外のtermですが、あまりに数が多くて全ての用語を確認していないのですが、一括して除外してしまうのもちょっと抵抗を感じます。
評価
50
翻訳 / 英語
- 2016/08/31 15:50:36に投稿されました
As first step, we decide a rule as follows, and matched scope automatically. As for group F, we decided that the term where there are more than 2 PRRs and number of report is more than 3 by any compound is ##.
Regarding remaining the lightest yellow, it overlaps work by Tomas. But we decided that the term only by name of inspection as well as the term whose negative and fake positive are shown are XX. For the terms other than that, as there are too many, we have not checked all. However, we feel uncomfortable if we exclude them all together.
Regarding remaining the lightest yellow, it overlaps work by Tomas. But we decided that the term only by name of inspection as well as the term whose negative and fake positive are shown are XX. For the terms other than that, as there are too many, we have not checked all. However, we feel uncomfortable if we exclude them all together.
評価
52
翻訳 / 英語
- 2016/08/31 15:52:08に投稿されました
As first step, I decide the rule below and attached the scope automatically. In group F, PRR is more than 2, and in each compound has 3 more reports. I name that as ##.
The rest of light color part, it duplicates what Tomas did but the word only used in inspection and negative, fake positive is named **. About other term, the amount is too numerous and haven't checked all the word but I feel reluctant to delete them all.
The rest of light color part, it duplicates what Tomas did but the word only used in inspection and negative, fake positive is named **. About other term, the amount is too numerous and haven't checked all the word but I feel reluctant to delete them all.
評価
51
翻訳 / 英語
- 2016/08/31 15:55:55に投稿されました
As the first step, we take a rule as following and define it in Scope mechanically.
In the group F, the terms which have more than 3 reports or 2 PRR in either compounds as ＃＃.
The rest of the weakest colors are the same as Tomas's work but only the terms with only inspections name and the terms with positive/negative implying are written as ＊＊.
And the terms other than above are too many to check and yet finished checking but don't feel right to remove them all.
In the group F, the terms which have more than 3 reports or 2 PRR in either compounds as ＃＃.
The rest of the weakest colors are the same as Tomas's work but only the terms with only inspections name and the terms with positive/negative implying are written as ＊＊.
And the terms other than above are too many to check and yet finished checking but don't feel right to remove them all.
我々の結果では、noiseが大きいものは認めませんでした。単に報告件数が少ないだけであれば、将来報告された時に適切に検出できるように残しておくべきかとも思いました。次にグループGです。色分けのルールはグループFと同じです。先程と異なり、原因が特定できない事象のグループなので、テスト結果のみで分類を決めてしまうのは困難です。全てのcompoundで報告件数が３以上になった用語は××と●●の２つですが、この２つは免疫抑制状態で生じる場合が多いですが、必ずしも疾患Aとは限りません。
評価
52
翻訳 / 英語
- 2016/08/31 15:58:10に投稿されました
In our result, we did not find noise that is loud. We thought that we should leave some to be adequately detected for the time when it is reported in the future if the reported number is only small. Next is group G. The color classification rule is the same as group F. The term which the reported number turned out three and more in every compound are two, ×× and ●●, however these two are likely to produce under immunosuppressed state and it is not necessarily disease A.
評価
50
翻訳 / 英語
- 2016/08/31 16:00:20に投稿されました
A large noise was not recognized by our result. If only the number of report is few, we should leave them to detect appropriately when it is reported in the future. Next is group G. The rule of classifying color is the same as that of group F. As it is a group of phenomenon we cannot specify the cause, which is different from before, it is difficult to decide classification only by result of test. The term where number of report is more than 3 by all compounds is xx and yy. These 2 occur when they contain being immune in many cases, but it is not always disease A.
評価
52
翻訳 / 英語
- 2016/08/31 15:58:20に投稿されました
In our result we couldn't find major noise. If the numbers of report is small, we should leave them so that we can properly detect in future when more report are submitted. Next is group G. The rule for classifying by color is same as group F. It is different from before, this is the group that the cause cannot be determined. So it is hard to classify by the results. The word that more than 3 numbers of all compounds report are XX and ●● but these two often occurs immunosuppressive state so it doesn't always mean disease A.
評価
50
翻訳 / 英語
- 2016/08/31 16:05:13に投稿されました
Our result did not recognize anything with large noise. If the number of report was simply not enough, I thought this should be remained so that we can detect it properly in case it is reported in the future. Next, let’s see the group-G. The same rule with the group-F was applied to color classification. Since it is a group with an event whose root cause cannot be identified unlike the previous one, it is difficult to determine classification only by this test result. The terms which had 3 or more in number of report were ×× and ●●, but these two are not necessarily Disease-A even though they are often generated with an immunocompromised state.
