Combined use of deep learning and CAE simulation for design and optimization of pharmaceutical products
| Project/Area Number |
17K08252
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Physical pharmacy
|
|---|
| Research Institution | Josai University |
|---|
Principal Investigator |
Takayama Kozo 城西大学, 薬学部, 招聘教授 (00130758)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
|
|---|
| Keywords | シミュレーション / 残留応力 / スパースモデリング / エラスティックネット / 深層学習 / 重要品質特性 / 錠剤形状 / X線CT / CAEシミュレーション / DPCモデル / 有限要素法 / 部分最小二乗法 / 製剤設計 / 錠剤 / 弾塑性体 / 引張強度 / 崩壊時間 / コンピュータ－シミュレーション / 密度分布 |
|---|
| Outline of Final Research Achievements |
Tablets are popular dosage forms for drug administration, and generally prepared by the compaction of pharmaceutical powders or granules so that various stresses can remain in the inside of tablets. The residual stresses may affect critical quality attributes (CQAs) of tablets, such as a hardness and a disintegration time. In this study, the residual stresses were estimated employing a finite element method incorporating a Drucker-Prager cap model. Hierarchical deep leaning models together with sparse models were applied to the remained stresses for predicting CQAs. As a result, CQAs were predicted with high accuracy as a function of remained stresses. The strength of residual stress was dependent in large part on the components of the formulations, but the distribution bias was mainly attributable to the tablet shape.
|
| Academic Significance and Societal Importance of the Research Achievements |
医薬品のおよそ50％は錠剤として投与されている。錠剤には，投与量を計数により調節できる，服用しやすい，携帯に便利である，安定性を担保しやすい等，多くの利点がある。しかし，その製法は複雑で，様々な処方成分と複数の工程を経て製造される。そのため，品質に優れる錠剤の製造には，現状では経験的要素が多く含まれる。本研究では，錠剤内部に残留する圧縮応力や引張応力に着目し，その強度と分布状態が，錠剤の重要な品質に及ぼす影響をAI技術を駆使することにより解明した。これより品質に優れる錠剤の設計を科学的根拠に基づいて実施することが可能となり，学術的意義に加え，社会的にも意義ある研究成果を得ることができた。
|
Report
(4 results)
Research Products
(31 results)
-
-
-
-
-
-
-
-
[Journal Article] Modeling of quantitative relationships between physicochemical properties of active pharmaceutical ingredients and tensile strength of tablets using a large dataset and a boosted tree2018
Author(s)
Hayashi, Y., Oishi, T., Shirotori, K., Marumo, Y., Kosugi, A., Kumada, S., Hirai, D., Takayama, K., Onuki, Y.
-
Journal Title
Drug Dev. Ind. Pharm.
Volume: in press
Issue: 7
Pages: 1090-1098
DOI
Related Report
Peer Reviewed / Open Access
-
-
-
-
[Journal Article] Relationships between response surfaces for tablet characteristics of placebo and API-containing tablets2017
Author(s)
Hayashi, Y., Tsuji, T., Shirotori, K., Oishi, T., Kosugi, A., Kumada, S., Hirai, D., Takayama, K., Onuki, Y.
-
Journal Title
Int. J. Pharm.
Volume: 532
Issue: 1
Pages: 82-89
DOI
Related Report
Peer Reviewed / Open Access
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
