Artificial Intelligence for solid tumor diagnosis in digital pathology

Role of artificial intelligence in digital pathology for gynecological cancers

The diagnosis of cancer is typically based on histopathological sections or biopsies on glass slides. Artificial intelligence (AI) approaches have greatly enhanced our ability to extract quantitative information from digital histopathology images as a rapid growth in oncology data. Gynecological cancers are major diseases affecting women's health worldwide. They are characterized by high mortality and poor prognosis, underscoring the critical importance of early detection, treatment, and identification of prognostic factors. This review highlights the various clinical applications of AI in gynecological cancers using digitized histopathology slides. Particularly, deep learning models have shown promise in accurately diagnosing, classifying histopathological subtypes, and predicting treatment response and prognosis. Furthermore, the integration with transcriptomics, proteomics, and other multi-omics techniques can provide valuable insights into the molecular features of diseases. Despite the considerable potential of AI, substantial challenges remain. Further improvements in data acquisition and model optimization are required, and the exploration of broader clinical applications, such as the biomarker discovery, need to be explored.

Computational pathology: A survey review and the way forward

Computational Pathology (CPath) is an interdisciplinary science that augments developments of computational approaches to analyze and model medical histopathology images. The main objective for CPath is to develop infrastructure and workflows of digital diagnostics as an assistive CAD system for clinical pathology, facilitating transformational changes in the diagnosis and treatment of cancer that are mainly address by CPath tools. With evergrowing developments in deep learning and computer vision algorithms, and the ease of the data flow from digital pathology, currently CPath is witnessing a paradigm shift. Despite the sheer volume of engineering and scientific works being introduced for cancer image analysis, there is still a considerable gap of adopting and integrating these algorithms in clinical practice. This raises a significant question regarding the direction and trends that are undertaken in CPath. In this article we provide a comprehensive review of more than 800 papers to address the challenges faced in problem design all-the-way to the application and implementation viewpoints. We have catalogued each paper into a model-card by examining the key works and challenges faced to layout the current landscape in CPath. We hope this helps the community to locate relevant works and facilitate understanding of the field's future directions. In a nutshell, we oversee the CPath developments in cycle of stages which are required to be cohesively linked together to address the challenges associated with such multidisciplinary science. We overview this cycle from different perspectives of data-centric, model-centric, and application-centric problems. We finally sketch remaining challenges and provide directions for future technical developments and clinical integration of CPath. For updated information on this survey review paper and accessing to the original model cards repository, please refer to GitHub. Updated version of this draft can also be found from arXiv.

Should sub-millimeter margins be deemed positive in oral cavity squamous cell carcinoma?

BACKGROUND

While several studies have indicated that a margin status of < 1 mm should be classified as a positive margin in oral cavity squamous cell carcinoma (OCSCC), there is a lack of extensive cohort studies comparing the clinical outcomes between patients with positive margins and margins < 1 mm.

METHODS

Between 2011 and 2020, we identified 18,416 Taiwanese OCSCC patients who underwent tumor resection and neck dissection. Of these, 311 had margins < 1 mm and 1013 had positive margins. To compare patients with margins < 1 mm and those with positive margins, a propensity score (PS)-matched analysis (n = 253 in each group) was conducted.

RESULTS

The group with margins < 1 mm displayed a notably higher prevalence of several variables: 1) tongue subsite, 2) younger age, 3) smaller depth of invasion), 4) early tumor stage, and 5) treatment with surgery alone. Patients with margins < 1 mm demonstrated significantly better disease-specific survival (DSS) and overall survival (OS) rates compared to those with positive margins (74 % versus 53 %, 65 % versus 43 %, both p < 0.0001). Multivariable analysis further confirmed that positive margins were an independent predictor of worse 5-year DSS (hazard ratio [HR] = 1.38, p = 0.0103) and OS (HR = 1.28, p = 0.0222). In the PS-matched cohort, the 5-year outcomes for patients with margins < 1 mm compared to positive margins were as follows: DSS, 71 % versus 59 %, respectively (p = 0.0127) and OS, 60 % versus 48 %, respectively (p = 0.0398).

CONCLUSIONS

OCSCC patients with a margin status < 1 mm exhibited distinct clinicopathological characteristics and a more favorable prognosis compared to those with positive resection margins.

A comprehensive survey of intestine histopathological image analysis using machine vision approaches

Colorectal Cancer (CRC) is currently one of the most common and deadly cancers. CRC is the third most common malignancy and the fourth leading cause of cancer death worldwide. It ranks as the second most frequent cause of cancer-related deaths in the United States and other developed countries. Histopathological images contain sufficient phenotypic information, they play an indispensable role in the diagnosis and treatment of CRC. In order to improve the objectivity and diagnostic efficiency for image analysis of intestinal histopathology, Computer-aided Diagnosis (CAD) methods based on machine learning (ML) are widely applied in image analysis of intestinal histopathology. In this investigation, we conduct a comprehensive study on recent ML-based methods for image analysis of intestinal histopathology. First, we discuss commonly used datasets from basic research studies with knowledge of intestinal histopathology relevant to medicine. Second, we introduce traditional ML methods commonly used in intestinal histopathology, as well as deep learning (DL) methods. Then, we provide a comprehensive review of the recent developments in ML methods for segmentation, classification, detection, and recognition, among others, for histopathological images of the intestine. Finally, the existing methods have been studied, and the application prospects of these methods in this field are given.

Department Wide Validation in Digital Pathology-Experience from an Academic Teaching Hospital Using the UK Royal College of Pathologists' Guidance

AIM

we describe our experience of validating departmental pathologists for digital pathology reporting, based on the UK Royal College of Pathologists (RCPath) "Best Practice Recommendations for Implementing Digital Pathology (DP)," at a large academic teaching hospital that scans 100% of its surgical workload. We focus on Stage 2 of validation (prospective experience) prior to full validation sign-off.

METHODS AND RESULTS

twenty histopathologists completed Stage 1 of the validation process and subsequently completed Stage 2 validation, prospectively reporting a total of 3777 cases covering eight specialities. All cases were initially viewed on digital whole slide images (WSI) with relevant parameters checked on glass slides, and discordances were reconciled before the case was signed out. Pathologists kept an electronic log of the cases, the preferred reporting modality used, and their experiences. At the end of each validation, a summary was compiled and reviewed with a mentor. This was submitted to the DP Steering Group who assessed the scope of cases and experience before sign-off for full validation. A total of 1.3% (49/3777) of the cases had a discordance between WSI and glass slides. A total of 61% (30/49) of the discordances were categorised as a minor error in a supplementary parameter without clinical impact. The most common reasons for diagnostic discordances across specialities included identification and grading of dysplasia, assessment of tumour invasion, identification of small prognostic or diagnostic objects, interpretation of immunohistochemistry/special stains, and mitotic count assessment. Pathologists showed similar mean diagnostic confidences (on Likert scale from 0 to 7) with a mean of 6.8 on digital and 6.9 on glass slide reporting.

CONCLUSION

we describe one of the first real-world experiences of a department-wide effort to implement, validate, and roll out digital pathology reporting by applying the RCPath Recommendations for Implementing DP. We have shown a very low rate of discordance between WSI and glass slides.

Deep Learning-Based Classification of Uterine Cervical and Endometrial Cancer Subtypes from Whole-Slide Histopathology Images

Uterine cervical and endometrial cancers have different subtypes with different clinical outcomes. Therefore, cancer subtyping is essential for proper treatment decisions. Furthermore, an endometrial and endocervical origin for an adenocarcinoma should also be distinguished. Although the discrimination can be helped with various immunohistochemical markers, there is no definitive marker. Therefore, we tested the feasibility of deep learning (DL)-based classification for the subtypes of cervical and endometrial cancers and the site of origin of adenocarcinomas from whole slide images (WSIs) of tissue slides. WSIs were split into 360 × 360-pixel image patches at 20× magnification for classification. Then, the average of patch classification results was used for the final classification. The area under the receiver operating characteristic curves (AUROCs) for the cervical and endometrial cancer classifiers were 0.977 and 0.944, respectively. The classifier for the origin of an adenocarcinoma yielded an AUROC of 0.939. These results clearly demonstrated the feasibility of DL-based classifiers for the discrimination of cancers from the cervix and uterus. We expect that the performance of the classifiers will be much enhanced with an accumulation of WSI data. Then, the information from the classifiers can be integrated with other data for more precise discrimination of cervical and endometrial cancers.

Deep learning for necrosis detection using canine perivascular wall tumour whole slide images

Necrosis seen in histopathology Whole Slide Images is a major criterion that contributes towards scoring tumour grade which then determines treatment options. However conventional manual assessment suffers from inter-operator reproducibility impacting grading precision. To address this, automatic necrosis detection using AI may be used to assess necrosis for final scoring that contributes towards the final clinical grade. Using deep learning AI, we describe a novel approach for automating necrosis detection in Whole Slide Images, tested on a canine Soft Tissue Sarcoma (cSTS) data set consisting of canine Perivascular Wall Tumours (cPWTs). A patch-based deep learning approach was developed where different variations of training a DenseNet-161 Convolutional Neural Network architecture were investigated as well as a stacking ensemble. An optimised DenseNet-161 with post-processing produced a hold-out test F1-score of 0.708 demonstrating state-of-the-art performance. This represents a novel first-time automated necrosis detection method in the cSTS domain as well specifically in detecting necrosis in cPWTs demonstrating a significant step forward in reproducible and reliable necrosis assessment for improving the precision of tumour grading.

[Oncological surgery in the interdisciplinary context-On the way to personalized medicine]

Oncological surgery is a discipline which closely interacts with other clinical partners and remains in many cases the cornerstone of a curative treatment of solid tumors. Due to the progress in the field of systemic tumor treatment as well as innovations in surgical techniques, the indications in oncological surgery are also changing, such as extended indications for patients with oligometastatic disease. Surgery of metastases has long been established for colorectal cancer and is being further tested for other entities, such as pancreatic and gastric cancer, within randomized controlled clinical trials (e.g. RENAISSANCE and METAPANC). A new challenge is the handling of a clinical complete remission after total neoadjuvant therapy, for example in locally advanced rectal cancer or in esophageal cancer. Here, organ and function preservation are increasingly propagated but should only be performed within clinical trials until stratification enables the identification of patients in whom this concept is oncologically safe. The personalized use of oncological surgery is dependent on the patient, the tumor and on the total multimodal concept.

Immunohistochemistry for Thymidine Kinase-1 (TK1): A Potential Tool for the Prognostic Stratification of Breast Cancer Patients

Breast cancer (BC) is the most frequent non-cutaneous malignancy in women. Histological grade, expression of estrogen and progesterone receptors (ER and PgR), overexpression/amplification of the human epidermal growth factor receptor 2 (HER2) oncogene, and proliferative activity measured with ki-67 provide important information on the biological features of BC and guide treatment choices. However, a biomarker that allows a more accurate prognostic stratification is still lacking. Thymidine kinase-1 (TK1), a ubiquitous enzyme involved in the pyrimidine nucleotide recovery pathway, is a cell-proliferation marker with potential prognostic and predictive impacts in BC. Eighty (80) cases of invasive BC with a long-term follow-up were retrospectively selected, and clinicopathological data were collected for each patient. TK1 tissue expression was evaluated immunohistochemically. Data suggested that TK1 expression levels are positively correlated with ER and PgR expression, and negatively correlated with HER2 status and the impact on patients’ distant recurrence-free survival (DRFS): in detail, among patients undergoing adjuvant chemotherapy, lower TK1 levels are correlated with better DRFS. Therefore, these results contribute to furthering the knowledge of TK1, suggesting a possible and important role of this enzyme as a biomarker in the stratification of BC patients.