Agenda

10:00 - 10:30

Check-In

10:30 - 10:40

Welcome Address

Hideya Makino

Country Manager

10:40 – 11:20

AI research at our hospital using HALO/HALO-AI

HALO/HALO-AIによる当院でのAI研究の取り組み

HALO/HALO-AI is a machine learning architecture built-in platform that can analyze digital pathology images as Whole Slide Imaging (WSI), allowing for a simple and intuitive workflow in digital pathology in a seamless manner. We are conducting research on quantitative and qualitative evaluation of lung cancer WSI data using HALO/HALO-AI in collaboration with the Department of Pathology, Laboratory, Respiratory Surgery, and Respiratory Medicine at our hospital. For quantitative evaluation, we are conducting the following studies: 1) quantification of lung cancer immune microenvironmental factors in relation to tumor mutation burden (TMB), 2) quantitative evaluation of tumor components, and 3) evaluation of a new evaluation index for preoperative treatment of lung cancer, on the other hand, qualitative evaluation includes (1) feasibility of ALK fusion gene prediction in lung cancer and (2) prediction of high TMB levels in lung cancer. In this talk, I would like to introduce our hospital's efforts with HALO/HALO-AI and also contribute to the discussion on the role of HALO/HALO-AI in the future development of medical AI from a clinician's perspective.

HALO/HALO-AIはデジタル病理画像をWhole Slide Imaging（WSI）のまま解析可能な機械学習アーキテクチャ内臓プラットフォームであり、デジタルパソロジーにおける簡便かつ直感的なワークフローをシームレスに行うことができる。当院では病理診断科・研究所・呼吸器外科・呼吸器内科共同にてHALO/HALO-AIによる肺癌WSIデータを用いた定量および定性評価に関する研究を行っている。定量評価に関しては①肺癌免疫微小環境因子の定量化とTMB（tumor mutation burden）との関連、②腫瘍構成成分の定量評価、③肺癌術前治療の新たな評価指標の評価について行い、一方で定性評価に関しては①肺癌ALK融合遺伝子予測の実現可能性、②肺癌TMB高値予測などを行っている。本ワークショップではHALO/HALO-AIによる当院の取り組み内容の紹介とともに、今後の医療AIが発展する中でのHALO/HALO-AIの役割についても臨床医の立場としてディスカッションに貢献したい。

Akira Ono, M.D., Ph.D.

Division of Thoracic Oncology

Shizuoka Cancer Center

小野　哲　先生

静岡県立静岡がんセンター

呼吸器内科　医長

11:20 – 12:00

Recognition and Quantification Study of Histopathological Changes in Pathological Animals Using HALO

HALOを用いた病態動物における病理組織変化の認識と定量化検討

Pathological examination in drug discovery is a powerful method to investigate the efficacy and toxicity of compounds by directly observing morphological changes in various tissues. With the recent development of digital pathology, it has become possible to recognize histopathological features and quantify the changes using AI as well. We have been working to improve the objectivity and accuracy of pathological evaluation and to streamline and save labor by using various image analysis software or collaborating with external institutions. Recently, we have attempted to quantify histopathological changes and findings using HALO, which have been difficult with conventional image analysis. In this talk, I will introduce the results of analysis of various pathological findings such as necrosis, atrophy, hyperplasia, basophilic changes, fibrosis, glomerular lesions, tumors, etc., using HALO and how to utilize the data obtained from the quantitative results. 

  Quantitative analysis of pathology often requires advanced expertise in setting up conditions and constructing analysis algorithms, in addition to pathological knowledge, but HALO is a user-friendly platform that is easy to use even for the beginners. Morphological recognition by HALO-AI has enabled analysis at the microscopic cellular level, and is expected to become a powerful support tool for pathological evaluation in the future.

創薬における病理検査は，様々な組織の形態学的変化を直接観察することにより，化合物の有効性や毒性を調べるのに有力な方法です．最近のデジタルパソロジーの発展に伴い，AIも利用した病理組織学的特徴を認識や，その変化の定量化が可能になってきている．我々はこれまで，様々な画像解析ソフトの使用あるいは外部機関との連携によって，病理評価の客観性・正確性の向上や作業の効率化・省力化に取り組んできた．最近では，HALOを使用して，従来の画像解析では困難であった病理組織学的変化や所見の定量化を試みている．この講演では，HALOを用いた様々な病理所見（壊死・萎縮・増生・好塩基性変化・線維化・糸球体病変・腫瘍など）の解析結果と定量結果から得られたデータの活用方法などに関して紹介する．

病理組織の定量的な解析には，病理学的な知識に加え，条件設定や解析アルゴリズム構築などの高度な専門知識が必要とされることが多いが，HALOはユーザーフレンドリーで初心者にも使い易いプラットフォームである．HALO-AIによる形態認識は，細胞単位のよりミクロなレベルでの解析も可能になってきており，今後も病理評価の強力なサポートツールとなることが期待される．

Yasushi Horai

Mitsubishi Tanabe Pharma Corporation 

洞井　康　様

田辺三菱製薬株式会社

12:00 - 12:50

Lunch

12:50 – 13:30

Histological Evaluation of Atrial Cardiomyopathy Using HALO/HALO-AI

HALO/HALO-AIを用いた心房心筋症の組織学的評価

Histological evaluation of atrial cardiomyopathy, the structural changes in the atrium underlying atrial fibrillation and heart failure, has not been studied to date because atrial specimens are difficult to obtain. We have developed a percutaneous catheter atrial biopsy technique and have performed 600 atrial biopsies to date. As a result, it was found that the atrial muscle of patients with atrial fibrillation had a wide variety of histological changes. Therefore, we performed various quantitative evaluations using HALO and histologically clarified the process of atrial degeneration (Takahashi, Yamaguchi, et al. Eur Heart J. 2023). Furthermore, we are currently developing a more objective and efficient quantitative evaluation method using HALO-AI. It is expected that detailed histological evaluation using HALO/HALO-AI will greatly advance the histological evaluation of cardiomyopathy in the future.

心房細動や心不全の背景に存在する心房の構造的変化、すなわち心房心筋症の組織学的評価は、心房検体の入手が困難であったことからこれまでほとんど研究されていなかった。我々は経皮的カテーテル心房生検術を開発し、これまで600例の心房生検を行った。その結果、心房細動症例の心房筋には非常に多彩な組織学的変化を有することが判明した。そこで、HALOを用いて様々な定量評価を行い、心房変性の過程を組織学的に明らかにした(Takahashi, Yamaguchi, et al. Eur Heart J. 2023)。さらに、現在ではHALO-AIを用いてより客観的かつ効率的な定量評価法を開発しつつある。HALO/HALO-AIを用いた詳細な組織学的評価により心筋症の組織学的評価は今後大きく前進することが期待される。

Takanori Yamaguchi

Associate Professor, Department of Cardiovascular Medicine

Saga University

山口　尊則　先生

佐賀大学

循環器内科　准教授

13:30 – 14:10

Indica Labs: Focusing on the Customer to Build for the Future

Indica Labs, Inc. was founded in 2011. Since that time, we have proudly focused on building tools that evolve with our customers’ image analysis and image management needs. By working to provide quality tools that fill a direct customer need, we have been able to build up a very large and very happy customer base, worldwide. As our company grows, so does the variety of Indica Labs, Inc.’s offerings. What started out as a strictly Research Use Only (RUO) company has now evolved to include Clinical products as well as Service offerings like our Pharma Services and Cloud Services teams. In Q1 2024, Indica Labs will be releasing v4.0 of HALO, HALO AI, and HALO Link. With our ever-growing product and service lines, our continued commitment to our original goal of evolving with the customer needs has stayed intact. This release will be our largest to date and will offer some very valuable new solutions to all of our end users, incorporating AI into more and more aspects of our products.

Adam Smith

VP, Life Science Applications - Global

Indica Labs 

14:10

Closing Remarks

Hideya Makino

Country Manager

14:30-16:00

Drinks Reception