A survey on artificial intelligence in histopathology image analysis

Abdelsamea, Mohammed M.; Zidan, Usama; Senousy, Zakaria; Gaber, Mohamed Medhat; Rakha, Emad A.; Ilyas, Mohammad

doi:10.1002/widm.1474

Cited by 29 publications

(14 citation statements)

References 190 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…AI has recently been shown to be useful in the analysis of medical images, 61 including angiograms 62 . Likewise, it has also been applied to histopathology image analysis workflow 63 . Our whole mount “vessel painting” angiograms are a reliable and simple combination of histology and angiography to evaluate the microvasculature 38,46,64,65 .…”

Section: Discussionmentioning

confidence: 99%

“…62 Likewise, it has also been applied to histopathology image analysis workflow. 63 Our whole mount "vessel painting" angiograms are a reliable and simple combination of histology and angiography to evaluate the microvasculature. 38,46,64,65 However, we speculated that the addition of AI could prove even more useful for their detailed analysis.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Vascular persistence following precision micropuncture

Horchler,

Hancock,

Sun

et al. 2023

Microcirculation

View full text Add to dashboard Cite

ObjectiveThe success of engineered tissues continues to be limited by time to vascularization and perfusion. Recently, we described a simple microsurgical approach, termed micropuncture (MP), which could be used to rapidly vascularize an adjacently placed scaffold from the recipient macrovasculature. Here we studied the long‐term persistence of the MP‐induced microvasculature.MethodsSegmental 60 μm diameter MPs were created in the recipient rat femoral artery and vein followed by coverage with a simple Type 1 collagen scaffold. The recipient vasculature and scaffold were then wrapped en bloc with a silicone sheet to isolate intrinsic vascularization. Scaffolds were harvested at 28 days post‐implantation for detailed analysis, including using a novel artificial intelligence (AI) approach.ResultsMP scaffolds demonstrated a sustained increase of vascular density compared to internal non‐MP control scaffolds (p < 0.05) secondary to increases in both vessel diameters (p < 0.05) and branch counts (p < 0.05). MP scaffolds also demonstrated statistically significant increases in red blood cell (RBC) perfused lumens.ConclusionsThis study further highlights that the intrinsic MP‐induced vasculature continues to persist long‐term. Its combination of rapid and stable angiogenesis represents a novel surgical platform for engineered scaffold and graft perfusion.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Vascular persistence following precision micropuncture

Horchler,

Hancock,

Sun

et al. 2023

Microcirculation

View full text Add to dashboard Cite

show abstract

“…Factors that should be considered arise when developing a DL model (71)(72)(73)(74)(75). Awareness of these factors is crucial to ensure a successful development process.…”

Section: Elements Impacting Model Developmentmentioning

confidence: 99%

Digital Pathology, Deep Learning, and Cancer: A Narrative Review

Williams,

Graifman,

Hussain

et al. 2024

Preprint

View full text Add to dashboard Cite

Background and Objective: Cancer is a leading cause of morbidity and mortality worldwide. The emergence of digital pathology and deep learning technologies signifies a transformative era in healthcare. These technologies can enhance cancer detection, streamline operations, and bolster patient care. A substantial gap exists between the development phase of deep learning models in controlled laboratory environments and their translations into clinical practice. This narrative review evaluates the current landscape of deep learning and digital pathology, analyzing the factors influencing model development and implementation into clinical practice. Methods: We searched multiple databases, including Web of Science, Arxiv, MedRxiv, BioRxiv, Embase, PubMed, DBLP, Google Scholar, IEEE Xplore, and Cochrane, targeting articles on whole slide imaging and deep learning published from 2014 and 2023. Out of 776 articles identified based on inclusion criteria, we selected 36 papers for the analysis. Key Content and Findings: Most articles in this review focus on the in-laboratory phase of deep learning model development, a critical stage in the deep learning lifecycle. Challenges arise during model development and their integration into clinical practice. Notably, lab performance metrics may not always match real-world clinical outcomes. As technology advances and regulations evolve, we expect more clinical trials to bridge this performance gap and validate deep learning models' effectiveness in clinical care. High clinical accuracy is vital for informed decision-making throughout a patient's cancer care. Conclusions: Deep learning technology can enhance cancer detection, clinical workflows, and patient care. Challenges may arise during model development. The deep learning lifecycle involves data preprocessing, model development, and clinical implementation. Achieving health equity requires including diverse patient groups and eliminating bias during implementation. While model development is integral, most articles focus on the pre-deployment phase. Future longitudinal studies are crucial for validating models in real-world settings post-deployment. A collaborative approach among computational pathologists, technologists, industry, and healthcare providers is essential for driving adoption in clinical settings. Keywords: Artificial Intelligence, Deep Learning, Digital Pathology, Computational Pathology, Cancer

show abstract

“…Hasil ekstraksi fitur dan klasifikasi dapat digunakan sebagai refrensi evaluasi pada teknik klasifikasi masa depan, namun penggunaan AlexNet belum pernah digunakan untuk klasifikasi citra RepoMedUNM. Pengolahan citra digital (Image processing) merupakan salah satu bidang ilmu dari Artificial Intelligence (AI) [11]. Proses image processing dengan teknik deteksi tepi merupakan proses yang digunakan untuk menghasilkan garis batas atau tepi-tepi dari objek citra untuk menandai bagian citra secara lebih detail dan memperbaiki detail citra yang kabur sehingga dapat menentukan jenis informasi awal terkait suatu penyakit [12].…”

Section: Pendahuluanunclassified

Klasifikasi Alexnet dan Deteksi Tepi Canny untuk Identifikasi Citra Repomedunm

Riana

Saputri

Hadianti

2023

jidt

View full text Add to dashboard Cite

Deteksi dini kanker serviks dapat mencegah dan menunda kematian, salah satunya dengan memanfaatkan teknologi komputer untuk mendiagnosa berbagai jenis sel kanker serviks. Penelitian dilakukan terhadap citra Pap smear yang diambil dari RepomedUNM dengan tujuan mengklasifikasikan citra Pap smear menjadi dua kelas yaitu sel normal dan sel abnormal dengan menggunakan metode AlexNet. Proses awal klasifikasi citra terdiri dari mengubah ukuran dan mengubah citra asli menjadi skala abu-abu. Penelitian ini juga bertujuan untuk mendeteksi tepi citra pap smear yang terdiri dari dua kelas yaitu sel normal dan sel koilocyt. Deteksi tepi menggunakan metode Canny untuk mendapatkan nilai luas, keliling dan diameter sel sitoplasma dan inti sel (nukleus). Proses deteksi tepi Canny terdiri dari proses cropping, mengubah citra asli menjadi grayscale, dan segmentasi citra menggunakan metode thresholding. Hasil klasifikasi 2000 citra Pap smear menghasilkan akurasi sebesar 97,66% dan hasil deteksi tepi dari 50 citra Pap smear dengan metode Canny mampu memberikan hasil yang baik dengan mendeteksi tepi citra sebenarnya dan hasilnya.

show abstract

A survey on artificial intelligence in histopathology image analysis

Cited by 29 publications

References 190 publications

Vascular persistence following precision micropuncture

Vascular persistence following precision micropuncture

Digital Pathology, Deep Learning, and Cancer: A Narrative Review

Klasifikasi Alexnet dan Deteksi Tepi Canny untuk Identifikasi Citra Repomedunm

Contact Info

Product

Resources

About