Contact us

    MrMrsMsDr



    Collection and Use of Personal Information. *

    Deep Bio’s AI Algorithm Helps Risk Stratification for Prostate Cancer

    Collaborative research by Deep Bio and Dr. Lotan research team at Johns Hopkins Medicine was revealed at AUA 2023

    SEOUL, SOUTH KOREA (PRWEB) MAY 11, 2023 – Deep Bio, a pioneer in medical AI for digital pathology and cancer diagnostics support software, announced that Dr. Lotan from Johns Hopkins Medicine presented their collaborative research which utilized the deep learning-based algorithm for prostate cancer diagnosis support at a podium session in the American Urological Association (AUA) annual meeting 2023.

    This joint research compared human pathologists and an AI algorithm in grading prostate biopsy specimens to predict biochemical recurrence after radical prostatectomy. Gleason score is one of the key elements to grade prostate cancer and can lead to different clinical decisions. For example, even though 3+4 and 4+3 Gleason scores are identical Gleason score 7s, each has a different prognosis. Gleason 3+4 requires more than active surveillance but tends to fare much better in prognosis than Gleason 4+3.

    The study reassessed the diagnoses of 284 patients initially diagnosed as Gleason grade group 2 (Gleason score 3+4) and underwent radical prostatectomy at Johns Hopkins Medicine from 2000 to 2014. This cohort is one of the most challenging cohorts to diagnose since many cases diagnosed 3+4 before could be downgraded to 3+3 or upgraded to 4+3. Johns Hopkins Medicine also has up to 14 years of follow-up data on the patients (an average of 4 years of follow-up) and approximately 16% of the patients went through recurrence of prostate cancer. As the ISUP guidelines have changed in those periods, all of these biopsies were re-graded by two expert genitourinary pathologists and a third expert genitourinary pathologist served as a tiebreak for discrepant reads. To compare human pathologists and AI algorithms, Deep Bio’s AI-based prostate cancer diagnosis support software DeepDx® Prostate analyzed all the biopsies to provide Gleason grades as well.

    The results showed poor agreements between the two pathologists, generating a kappa of 0.17 while grading between the consensus pathology read and the AI algorithm generated a slightly higher kappa of 0.33. Importantly, however, the study suggested that the algorithm can act as a tool that stratifies patients for subsequent biochemical recurrence after radical prostatectomy. This risk stratification has the potential to prevent a lot of unnecessary surgeries and aid in choosing between treatment modalities.

    “It is a meaningful milestone for Deep Bio to conduct a joint research with Johns Hopkins Medicine, one of the top global universities, and the findings are remarkable for prostate cancer diagnosis using AI,” said Sun-Woo Kim, CEO of Deep Bio. “As precise diagnosis and personalization have become the crucial factors in deciding medical decisions for medical professionals, we will focus on bringing innovative deep learning technologies that can unlock new insights in cancer diagnosis and treatment” added he.

    Deep Bio continues to focus on not only AI diagnostics but also R&D and presenting the results in various international conferences and events. The company also continues to build its presence in the global market through collaboration with overseas digital pathology solution providers in the US, Europe, and India.

    About Deep Bio
    Deep Bio Inc. is an AI healthcare company with in-house expertise in deep learning and cancer pathology. Our vision is to radically improve efficiency and accuracy of pathologic cancer diagnosis and prognosis, by equipping pathologists with deep learning-based IVD SaMDs (In Vitro Diagnostics Software as a Medical Device), for optimal cancer treatment decisions. To learn more, visit http://www.deepbio.co.kr.

    DeepDx® Prostate is a clinically-validated AI for prostate core needle biopsy tissue image analysis. Whole-slide images (WSIs) of H&E-stained biopsy tissue specimens are analyzed for prostate cancer, Gleason scores and grade groups. Extensively tested at 4 US CLIA labs (700k+ cores between 2019 and 2021), DeepDx® Prostate can alleviate the shortage of pathologists and the resultant increase in workload, while reducing diagnostic subjectivity and variability. To learn more, visit http://www.deepbio.co.kr

    [Abstract] ASCO 2023 – Validation of AI-based postoperative nomograms for biochemical recurrence in prostate adenocarcinoma

    Authors:
    JaeHeon Lee, Tae-Yeong Kwak, Joonyoung Cho, Sun Woo Kim, Hyeyoon Chang, Hong Koo Ha

    Organizations
    Deep Bio Inc., Seoul, South Korea, Pusan National University, Pusan, South Korea

    Background
    After radical prostatectomy (RP), a steep increase in PSA level is an early sign of the disease progression in prostate cancer, which is known as biochemical recurrence (BCR). The risk of BCR can be evaluated based on a combination of clinicopathological factors, and the patient’s Gleason score plays a significant role. However, this scoring system may have lower consistency due to interobserver reproducibility in classifying Gleason Patterns (GP) as well as in quantifying the amount of each GP. We developed AI-based nomograms, with the aim of investigating their prognostic efficacy.

    Methods
    In this study, digitized whole-slide images (WSIs) of H&E-stained prostatectomy specimens and clinical follow-up information were obtained from two sources: Pusan National University Hospital (PNUH, n = 967, event = 342) from 2010 to 2021 with the median follow-up being 3.7 years, and The Cancer Genome Atlas (TCGA, n = 352, event = 79) from 2000 to 2013 with the median follow-up 2.6 years. We used the DeepDx Prostate – RP, an AI-based prostate cancer Gleason grading model, to compute a pixel-wise probability map of each GP in WSIs. Then, a weighted sum of the probabilities and GPs was calculated for each pixel. The proposed slide-level score (AI score) was then determined by averaging them across all pixels in WSI, resulting in a value ranging from 3 to 5. Also, patients were divided into five groups using AI score thresholds (3.1, 3.5, 3.9, and 4.1), and five binary variables (AI score group 1-5) were generated, where a value of 1 indicates a patient’s categorization into a group. To evaluate predictability we created new nomograms incorporating AI scores based on existing nomograms: MSKCC and CAPRA-S. Unlike the original nomograms, the proposed nomograms do not include the Grade Group (GG) made by pathologists. Then, we fitted a Cox regression model on one of two datasets using the original and newly formed nomograms, and validated on the other dataset reporting the concordance index (c-index). Confidence intervals for the c-index were generated via the non-parametric bootstrap resampling with 9,999 samples.

    Results
    The table shows the prognostic performance of sets of clinicopathological factors and demonstrates that the proposed AI score improved the predictive power. The nomograms with AI score group 1-5 outperformed the original nomograms, as shown in (d).

    Conclusions
    In conclusion, we developed the AI-based nomograms which can improve the accuracy of predicting the biochemical recurrence in prostate cancer compared to the existing nomograms.

    [Abstract] ASCO 2023 – Deep learning-based histomorphological pattern profiles for effective risk stratification in prostate cancer.

    Authors:
    JaeHeon Lee, Tae-Yeong Kwak, Joonyoung Cho, Sun Woo Kim, Hyeyoon Chang

    Organizations
    Deep Bio Inc., Seoul, South Korea

    Background
    Prostate cancer affects millions of men globally and biochemical recurrence (BCR) is an important indicator of its progression. With the increasing use of computational techniques, there has been a growing interest in utilizing whole slide images (WSIs) of H&E stained prostate tissue to evaluate the patients’ clinical risks. Our study aimed to develop a novel method for predicting BCR by performing clustering on features extracted from deep learning models, thereby capturing morphological characteristics.

    Methods
    The WSIs and clinical information were collected from The Cancer Genome Atlas (TCGA, from 2000 to 2013) dataset, for 321 patients (43 patients with BCR) who were not treated with any adjuvant therapies, with 2.5 years of median follow-up time. 356 collected WSIs were split into 1024 x 1024 patches at 5x magnification for the deep learning-based analysis. We used an ImageNet-pretrained deep learning-based clustering model (SwAV) to extract features from each patch and applied additional K-means clustering to assign the patches into one of 16 clusters. We tested both the cosine similarity and Euclidean distance as the clustering measure. After excluding the dissimilar patches under specific thresholds, the proportion of each cluster was calculated per slide to describe the distribution of morphological patterns as a 16-dimensional vector. To evaluate the predictability, we trained Cox models with the proportion vectors and assessed their performance via the concordance index (c-index). We repeated a stratified 5-fold cross-validation process 100 times to achieve robustness. As a reference, we conducted the same experiment with a BCR prediction nomogram published by MSKCC. Moreover, with the median proportion of each cluster, we separated two patient groups, drawed Kaplan-Meier curves, performed the log-rank test for significance, and we obtained the hazard ratio (HR) of each cluster from the trained Cox model of each fold.

    Results
    We explored several thresholds to find ones yielding the best performance. When using the similarity, the best c-index was 0.704 with the threshold value of 0.25. With the distance value of 0.60, the c-index was 0.701. When using the MSKCC nomogram that incorporates clinical features, the average c-index was 0.717. The table contains the results of statistical analysis for selected clusterings that were significant in the log-rank test. Particularly in cluster B, the patches showed high-risk patterns including perineural invasion, high grade cancer, and necrosis in agreement with the high HR.

    Conclusions
    Our results suggest that the histomorphological pattern profiles depicted by the deep learning model have the potential for effective risk stratification in prostate cancer.

    [Abstract] ASCO 2023 – Exploring the efficacy of a continuous form of the histologic grade in prostate cancer prognosis prediction

    Authors:
    Tae-Yeong Kwak, JaeHeon Lee, Joonyoung Cho, Sun Woo Kim, Hyeyoon Chang

    Organizations
    Deep Bio Inc., Seoul, South Korea

    Background

    Adenocarcinoma of the prostate is the second most common type of cancer among men worldwide. The Gleason grading system remains the gold standard for evaluating the prognosis of prostate cancer by assessing the cancer morphology. However, the current discretized Gleason pattern regime limits the depiction of fine-grained histomorphological changes. We evaluated the efficacy of the algorithm-generated continuous histologic grade value in prostate cancer prognosis prediction.

    Methods
    Whole-slide images of H&E prostatectomy tissue, along with the follow-up data including new tumor event (NTE) and biochemical recurrence (BCR), were obtained for the cases diagnosed as prostate cancer during 2000-2013 from The Cancer Genome Atlas (TCGA) database. Cases with missing data were excluded, resulting in a total of 308 and 397 cases being used for the analysis of NTE and BCR, respectively. Our study utilized a deep learning-based algorithm that performs Gleason scoring as follows. First, it computes per-pixel likelihood values for each of the 4 classes: benign, Gleason 3, 4, and 5. Then, the per-pixel class with the highest likelihood value and the slide-wise ISUP grade group (GG-AI) is successively determined. We tweaked the algorithm to aggregate the likelihood-weighted Gleason grade assigned to each pixel into a continuous form of the histologic grade (c-HG). We compared the prognostic performance of c-HG with that of the original ISUP grade group in TCGA (GG) and that of algorithm-generated GG-AI in predicting the risk of NTE as well as BCR. The Cox regression analysis was conducted, setting the cases from one of three donating institutions (EJ, HC, KK) as the target, fitting the Cox model with the cases from the other institutions, and evaluating the fitted model on the target.

    Results
    The median follow-up duration in months was 32 for NTE and 29 for BCR, respectively. The number of cases with NTE was 51, while the one with BCR was 82. The table presents the c-index values obtained from each institution for GG, GG-AI, and c-HG. Note that the number of cases in each institution is different for NTE and BCR, due to the different number of removed cases with missing data. c-HG showed the best average performance in both NTE and BCR risk predictions. It is also notable that c-HG showed a stable performance for varying institutions.

    Conclusions
    Our findings support the usefulness of the algorithm-based analysis of prostatectomy specimens, which can benefit clinicians in the hospital.

    We proposed an algorithm-based method of representing histologic grade as a continuous value, which may give better predictions in disease progression for prostate adenocarcinoma.

    [Abstract] ASCO 2023 – Algorithm-based histologic grade and tumor ratio for radical prostatectomy: Comparison with pathology reports.

    Authors:
    Tae-Yeong Kwak, JaeHeon Lee, Joonyoung Cho, Sun Woo Kim, Hyeyoon Chang, Hong Koo Ha

    Organizations
    Deep Bio Inc., Seoul, South Korea, Pusan National University, Pusan, South Korea

    Background

    The histologic grade (Gleason score) as well as the tumor ratio of the resected specimen plays a significant role in assessing the clinical risk of prostate cancer patients who underwent radical prostatectomy. Gleason scoring is known to be prone to inter- and intra-observer discordance, and eyeballing measurement of tumor ratio is possibly coarse and inaccurate. Several deep learning-based tissue image analysis algorithms that perform prostate cancer diagnosis have been developed and are about to be applied clinically. We evaluated the utility of one of those algorithms by comparing its output with the pathology reports.

    Methods
    A total of 29681 H&E-stained tissue slides were collected for 1001 radical prostatectomy cases during 2010-2021 at Pusan National University Hospital and scanned at 40x magnification into whole-slide images. On each slide, we utilized a tissue detection algorithm to identify tissue regions and measure the area. A deep learning-based algorithm was used to identify prostate cancer lesions, measure the area, and determine the grades. The area measurement was converted into the volume figure assuming that the tissue slice thickness was 5mm. The slide-wise algorithm outputs were then aggregated for each case, resulting in the specimen volume, tumor ratio, and Gleason score. In the evaluation, the algorithm-based Gleason scores were compared with the ones in the hospital pathology reports on the ISUP grade group basis. The correlation analysis was performed for the tumor ratio. The specimen density values were calculated from the volume figures and the hospital weight measurement and utilized to exclude the cases with extreme density values.

    Results
    The min, max, and average number of slides per case were 1, 67, and 29.7. For two cases with a single slide collected, both the algorithm and the expert pathologist found no cancer. The pathologist also confirmed that 7 had residual tumors among 15 cases where there was no Gleason score in the pathology reports while the algorithm found cancer. For the remaining 984 cases, the algorithm provided the grade groups equal to, higher than, and lower than the pathology reports for 482, 384, and 118 cases respectively, showing moderate agreement. The min, max, and mean values of the specimen volume (mL) were 0.01, 75.3, and 32.7, while the corresponding values of the specimen density (g/mL) were 0.11, 3001.79, and 4.38. The density values at the lower and upper 2.5% were 0.42 and 1.40 respectively, and the mean of the values between them (inner 95%) was 0.92. The degree of tumor ratio correlation between the algorithm and the pathology reports was high with the coefficient 0.813 (95% CI: 0.791-0.833), which went up to 0.826 (95% CI: 0.805-0.845) for the inner 95% density cases.

    Conclusions
    Our findings support the usefulness of the algorithm-based analysis of prostatectomy specimens, which can benefit clinicians in the hospital.