A novel dataset and efficient deep learning framework for automated grading of renal cell carcinoma from kidney histopathology images

Trends of kidney cancer cases worldwide are expected to increase persistently and this inspires the modification of the traditional diagnosis system to respond to future challenges. Renal Cell Carcinoma (RCC) is the most common kidney cancer and responsible for 80-85% of all renal tumors. This study proposed a robust and computationally efficient fully automated Renal Cell Carcinoma Grading Network (RCCGNet) from kidney histopathology images. The proposed RCCGNet contains a shared channel residual (SCR) block which allows the network to learn feature maps associated with different versions of the input with two parallel paths. The SCR block shares the information between two different layers and operates the shared data separately by providing beneficial supplements to each other. As a part of this study, we also introduced a new dataset for the grading of RCC with five different grades. We obtained 722 Hematoxylin & Eosin (H &E) stained slides of different patients and associated grades from the Department of Pathology, Kasturba Medical College (KMC), Mangalore, India. We performed comparable experiments which include deep learning models trained from scratch as well as transfer learning techniques using pre-trained weights of the ImageNet. To show the proposed model is generalized and independent of the dataset, we experimented with one additional well-established data called BreakHis dataset for eight class-classification. The experimental result shows that proposed RCCGNet is superior in comparison with the eight most recent classification methods on the proposed dataset as well as BreakHis dataset in terms of prediction accuracy and computational complexity.

Appendiceal neuroendocrine neoplasms in children and adolescents – Two case reports from a tertiary care center in coastal India

Appendiceal neuroendocrine neoplasms (ANENs) are uncommon in children and adolescents and are incidentally diagnosed in 0.3% of appendicectomy specimens. Pediatric ANENs rarely metastasize and have excellent prognosis. We report two cases of ANENs in adolescents who presented clinically and radiologically with features of acute appendicitis for which they underwent appendicectomy. On gross examination, £ 2 cm lesion was identified in both the appendix. Microscopic and immunohistochemical analysis clinched the diagnosis of ANENs. Patients were followed up for 18 months and were disease free. This report emphasizes that a possibility of ANEN should be kept as a differential diagnosis even in the pediatric population presenting with acute appendicitis. A routine histopathological examination of all appendicectomy specimens is therefore crucial, as early diagnosis is associated with excellent prognosis.

Deep structured residual encoder-decoder network with a novel loss function for nuclei segmentation of kidney and breast histopathology images

To improve the process of diagnosis and treatment of cancer disease, automatic segmentation of haematoxylin and eosin (H & E) stained cell nuclei from histopathology images is the first step in digital pathology. The proposed deep structured residual encoder-decoder network (DSREDN) focuses on two aspects: first, it effectively utilized residual connections throughout the network and provides a wide and deep encoder-decoder path, which results to capture relevant context and more localized features. Second, vanished boundary of detected nuclei is addressed by proposing an efficient loss function that better train our proposed model and reduces the false prediction which is undesirable especially in healthcare applications. The proposed architecture experimented on three different publicly available H&E stained histopathological datasets namely: (I) Kidney (RCC) (II) Triple Negative Breast Cancer (TNBC) (III) MoNuSeg-2018. We have considered F1-score, Aggregated Jaccard Index (AJI), the total number of parameters, and FLOPs (Floating point operations), which are mostly preferred performance measure metrics for comparison of nuclei segmentation. The evaluated score of nuclei segmentation indicated that the proposed architecture achieved a considerable margin over five state-of-the-art deep learning models on three different histopathology datasets. Visual segmentation results show that the proposed DSREDN model accurately segment the nuclear regions than those of the state-of-the-art methods.

High-resolution deep transferred ASPPU-Net for nuclei segmentation of histopathology images

PURPOSE

Increasing cancer disease incidence worldwide has become a major public health issue. Manual histopathological analysis is a common diagnostic method for cancer detection. Due to the complex structure and wide variability in the texture of histopathology images, it has been challenging for pathologists to diagnose manually those images. Automatic segmentation of histopathology images to diagnose cancer disease is a continuous exploration field in recent times. Segmentation and analysis for diagnosis of histopathology images by using an efficient deep learning algorithm are the purpose of the proposed method.

METHOD

To improve the segmentation performance, we proposed a deep learning framework that consists of a high-resolution encoder path, an atrous spatial pyramid pooling bottleneck module, and a powerful decoder. Compared to the benchmark segmentation models having a deep and thin path, our network is wide and deep that effectively leverages the strength of residual learning as well as encoder-decoder architecture.

RESULTS

We performed careful experimentation and analysis on three publically available datasets namely kidney dataset, Triple Negative Breast Cancer (TNBC) dataset, and MoNuSeg histopathology image dataset. We have used the two most preferred performance metrics called F1 score and aggregated Jaccard index (AJI) to evaluate the performance of the proposed model. The measured values of F1 score and AJI score are (0.9684, 0.9394), (0.8419, 0.7282), and (0.8344, 0.7169) on the kidney dataset, TNBC histopathology dataset, and MoNuSeg dataset, respectively.

CONCLUSION

Our proposed method yields better results as compared to benchmark segmentation methods on three histopathology datasets. Visual segmentation results justify the high value of the F1 score and AJI scores which indicated that it is a very good prediction by our proposed model.

Efficient deep learning architecture with dimension-wise pyramid pooling for nuclei segmentation of histopathology images

Image segmentation remains to be one of the most vital tasks in the area of computer vision and more so in the case of medical image processing. Image segmentation quality is the main metric that is often considered with memory and computation efficiency overlooked, limiting the use of power hungry models for practical use. In this paper, we propose a novel framework (Kidney-SegNet) that combines the effectiveness of an attention based encoder-decoder architecture with atrous spatial pyramid pooling with highly efficient dimension-wise convolutions. The segmentation results of the proposed Kidney-SegNet architecture have been shown to outperform existing state-of-the-art deep learning methods by evaluating them on two publicly available kidney and TNBC breast H&E stained histopathology image datasets. Further, our simulation experiments also reveal that the computational complexity and memory requirement of our proposed architecture is very efficient compared to existing deep learning state-of-the-art methods for the task of nuclei segmentation of H&E stained histopathology images. The source code of our implementation will be available at https://github.com/Aaatresh/Kidney-SegNet.

LiverNet: efficient and robust deep learning model for automatic diagnosis of sub-types of liver hepatocellular carcinoma cancer from H&E stained liver histopathology images

PURPOSE

Liver cancer is one of the most common types of cancers in Asia with a high mortality rate. A common method for liver cancer diagnosis is the manual examination of histopathology images. Due to its laborious nature, we focus on alternate deep learning methods for automatic diagnosis, providing significant advantages over manual methods. In this paper, we propose a novel deep learning framework to perform multi-class cancer classification of liver hepatocellular carcinoma (HCC) tumor histopathology images which shows improvements in inference speed and classification quality over other competitive methods.

METHOD

The BreastNet architecture proposed by Togacar et al. shows great promise in using convolutional block attention modules (CBAM) for effective cancer classification in H&E stained breast histopathology images. As part of our experiments with this framework, we have studied the addition of atrous spatial pyramid pooling (ASPP) blocks to effectively capture multi-scale features in H&E stained liver histopathology data. We classify liver histopathology data into four classes, namely the non-cancerous class, low sub-type liver HCC tumor, medium sub-type liver HCC tumor, and high sub-type liver HCC tumor. To prove the robustness and efficacy of our models, we have shown results for two liver histopathology datasets-a novel KMC dataset and the TCGA dataset.

RESULTS

Our proposed architecture outperforms state-of-the-art architectures for multi-class cancer classification of HCC histopathology images, not just in terms of quality of classification, but also in computational efficiency on the novel proposed KMC liver data and the publicly available TCGA-LIHC dataset. We have considered precision, recall, F1-score, intersection over union (IoU), accuracy, number of parameters, and FLOPs as metrics for comparison. The results of our meticulous experiments have shown improved classification performance along with added efficiency. LiverNet has been observed to outperform all other frameworks in all metrics under comparison with an approximate improvement of [Formula: see text] in accuracy and F1-score on the KMC and TCGA-LIHC datasets.

CONCLUSION

To the best of our knowledge, our work is among the first to provide concrete proof and demonstrate results for a successful deep learning architecture to handle multi-class HCC histopathology image classification among various sub-types of liver HCC tumor. Our method shows a high accuracy of [Formula: see text] on the proposed KMC liver dataset requiring only 0.5739 million parameters and 1.1934 million floating point operations per second.

NucleiSegNet: Robust deep learning architecture for the nuclei segmentation of liver cancer histopathology images

The nuclei segmentation of hematoxylin and eosin (H&E) stained histopathology images is an important prerequisite in designing a computer-aided diagnostics (CAD) system for cancer diagnosis and prognosis. Automated nuclei segmentation methods enable the qualitative and quantitative analysis of tens of thousands of nuclei within H&E stained histopathology images. However, a major challenge during nuclei segmentation is the segmentation of variable sized, touching nuclei. To address this challenge, we present NucleiSegNet - a robust deep learning network architecture for the nuclei segmentation of H&E stained liver cancer histopathology images. Our proposed architecture includes three blocks: a robust residual block, a bottleneck block, and an attention decoder block. The robust residual block is a newly proposed block for the efficient extraction of high-level semantic maps. The attention decoder block uses a new attention mechanism for efficient object localization, and it improves the proposed architecture's performance by reducing false positives. When applied to nuclei segmentation tasks, the proposed deep-learning architecture yielded superior results compared to state-of-the-art nuclei segmentation methods. We applied our proposed deep learning architecture for nuclei segmentation to a set of H&E stained histopathology images from two datasets, and our comprehensive results show that our proposed architecture outperforms state-of-the-art methods. As part of this work, we also introduced a new liver dataset (KMC liver dataset) of H&E stained liver cancer histopathology image tiles, containing 80 images with annotated nuclei procured from Kasturba Medical College (KMC), Mangalore, Manipal Academy of Higher Education (MAHE), Manipal, Karnataka, India. The proposed model's source code is available at https://github.com/shyamfec/NucleiSegNet.

Analysis of Dedifferentiated Liposarcomas Emphasizing the Diagnostic Dilemmas

Introduction: Dedifferentiated liposarcoma (DDLPS) is defined as a progression of well-differentiated liposarcoma (WDLPS) to another nonlipogenic sarcoma. Since a variety of heterologous sarcomas can be encountered in dedifferentiation, clinical dilemmas arise. The present study analyzed the role of clinicopathologic and immunohistochemical (IHC) features in the diagnosis of DDLPS and its differentiation from mimics. Materials and Methods: A retrospective and prospective study was conducted wherein all cases of liposarcoma from 2012 to 2017 were reviewed. DDLPS cases were identified among pleomorphic lesions. Clinical and histopathological details for these cases were retrieved from medical records section and department archives. Histomorphology and immunohistochemistry (MDM2, S100, and Ki-67) were analyzed for these cases. Results: Among 37 cases of liposarcomas reviewed, DDLPS was diagnosed in 12 cases (32.4%). Mean age of the patients was 54.3 years with equal gender distribution (M:F =1:1.2). Two patients had recurrent tumors. Most were retroperitoneal (58.3%) with mean duration of symptoms being 8.7 months. Mean tumor dimension was 17.5 cm. High-grade dedifferentiated component was most common (83.3%) with only one case each (8.3%) of low-grade and homologous dedifferentiation. Undifferentiated pleomorphic sarcoma was the frequent nonlipogenic sarcoma. MDM2 overexpression was detected in 100%, focal S100 positivity seen in 66.6%, and mean Ki-67 labeling index was 24. Conclusion: DDLPS exhibits aggressive clinical behavior. Adequate sampling, correlation to clinical details, demonstration of transition from WDLPS to DDLPS aid in narrowing the differentials. Immunostaining with MDM2 helps in definite categorization and S100 highlights lipoblasts, when they are not easily identifiable. MDM2, CDK4, and p16 IHC panel is recommended in all cases and fluorescence in situ hybridization analysis where IHC is noncontributory.

A study about color normalization methods for histopathology images

Histopathology images are used for the diagnosis of the cancerous disease by the examination of tissue with the help of Whole Slide Imaging (WSI) scanner. A decision support system works well by the analysis of the histopathology images but a lot of problems arise in its decision. Color variation in the histopathology images is occurring due to use of the different scanner, use of various equipments, different stain coloring and reactivity from a different manufacturer. In this paper, detailed study and performance evaluation of color normalization methods on histopathology image datasets are presented. Color normalization of the source image by transferring the mean color of the target image in the source image and also to separate stain present in the source image. Stain separation and color normalization of the histopathology images can be helped for both pathology and computerized decision support system. Quality performances of different color normalization methods are evaluated and compared in terms of quaternion structure similarity index matrix (QSSIM), structure similarity index matrix (SSIM) and Pearson correlation coefficient (PCC) on various histopathology image datasets. Our experimental analysis suggests that structure-preserving color normalization (SPCN) provides better qualitatively and qualitatively results in comparison to the all the presented methods for breast and colorectal cancer histopathology image datasets.

The Role of Crush Cytology in the Diagnosis of Large-Intestine Lesions with Correlation on Histopathology

OBJECTIVE

To study the efficacy of colonoscopic crush cytology as a convenient and near-accurate method to evaluate colonic neoplasms.

STUDY DESIGN

Retrospective and cross-sectional. The original cytologic diagnoses were correlated with a histology report on 100 cases sent to the cytology laboratory over 2 years.

RESULTS

Of the 100 cases, 25 were nonmalignant. Of the 75 malignant lesions, 72 could be identified as positive for malignancy on cytology. The false-positives consisted of 6 adenomas and 1 case of ulcerative colitis. Thus, sensitivity and specificity of cytology are 96 and 63.2%, respectively. Of the 6 adenomas diagnosed as malignant, 4 showed high-grade dysplasia, and the other 2 showed superficial ulceration with low-grade dysplasia on histopathology. The ulcerative colitis case showed widespread ulcers and regenerative/reparative features on biopsy. The 3 adenocarcinomas diagnosed s benign on cytology showed an occasional malignant cell with thickened nuclear borders and prominent central nucleoli.

CONCLUSIONS

With careful attention to the cytomorphology, coupled with good clinical and endoscopic correlation, crush cytology of the large intestine is a reliable diagnostic tool. It categorizes lesions as malignant and benign with a high sensitivity, positive predictive value, and negative predictive value. Adenomas and reparative/regenerative changes seen in inflammatory bowel disease are major pitfalls in the cytology diagnosis of malignancy that may be averted by informing the endoscopic findings and clinical history. Cytology diagnosis saves time and gives proper feedback to the gastroenterologist.

The role of discriminant functions in screening beta thalassemia trait and iron deficiency anemia among laboratory samples

INTRODUCTION

Most important differential diagnosis for microcytosis and hypochromia is beta thalassemia trait (BTT) and iron deficiency anemia.

AIM

To study the utility of discriminant functions (DFs) and red cell indices in distinguishing BTT and iron deficiency anemia.

METHODS

The study is observational (cross sectional). A total of 350 patients, 43 BTT, and 307 iron-deficiency anemia reflecting actual disease prevalence were included. Their complete red blood cell parameters, hemoglobin A2, and serum ferritin level wherever required were obtained. Receiver operator characteristic curve was drawn for each DF and results compared with other studies.

RESULTS

Among the six DFs, the highest sensitivity (97.7%) and specificity (98.6%) was shown, respectively, by Shine and Lal (S and L) and England and Fraser index (E and F) in identifying cases of BTT. Youden index of the Mentzer index (MI) was the highest (69.0) and S and L, the lowest (13.2) indicating MI to be the most efficient and the S and L, the least in differentiating the two entities. Red cell distribution width index (RDWI) showed the highest accuracy (91.6%), whereas S and L showed the least accuracy (29.6%).

CONCLUSION

MI was the most efficient in discriminating BTT from iron deficiency anemia (IDA). RDWI stands to be the most accurate. S and L could at best be used as screening tool rather than DF. No study except one agreed with us because convenient sampling used in other studies generated bias in their results. Statistically, this study bears far more relevance than other studies because the sample distribution reflects the prevalence of IDA and BTT in the community.

Disseminated histoplasmosis with conjunctival involvement in an immunocompromised patient

We report a case of disseminated histoplasmosis in a 37-year-old male acquired immunodeficiency syndrome patient from south India. The patient presented with high-grade fever, cough, conjunctival nodule and papulonodular hyperpigmented skin lesions. Histology of skin lesions and conjunctival nodule showed numerous intracellular Periodic Acid Schiff-positive rounded yeast cells within macrophages. Bone marrow aspirate confirmed disseminated histoplasmosis. The patient showed dramatic response after starting treatment with Amphotercin B.

Disseminated histoplasmosis with conjunctival involvement in an immunocompromised patient

We report a case of disseminated histoplasmosis in a 37-year-old male acquired immunodeficiency syndrome patient from south India. The patient presented with high-grade fever, cough, conjunctival nodule and papulonodular hyperpigmented skin lesions. Histology of skin lesions and conjunctival nodule showed numerous intracellular Periodic Acid Schiff-positive rounded yeast cells within macrophages. Bone marrow aspirate confirmed disseminated histoplasmosis. The patient showed dramatic response after starting treatment with Amphotercin B.

Mechanism delineating differential effect of an antiestrogen, tamoxifen, on the serum LH and FSH in adult male rats

Tamoxifen, a synthetic non-steroidal antiestrogen with residual estrogenic activity, administered to adult male rats reduces their fertility. A decrease in the circulating LH and testosterone levels with a transient rise or no change in circulating FSH levels was observed. The present study was carried out to delineate the mechanism causing the differential effect of tamoxifen on circulating gonadotropins by correlating it to changes in the hypothalamic LHRH, pituitary gonadotropins and testicular inhibin/activin. Hypothalamus, pituitary-hypothalamus complex (PHC) and intact pituitary (PI) from control and tamoxifen-treated male rats were superfused in vitro, and pulsatile release of LHRH by hypothalamus and that of LH and FSH by the PHC and PI were studied. Concomitantly, testicular immunoexpression of alpha and betaB subunits of inhibin/activin were studied by immunohistochemistry and enzyme-linked immunosorbent assays (ELISA). At 0.4 mg/kg/day dose of tamoxifen a decrease in mean hypothalamic LHRH and LH pulse frequency from PHC construct was observed. FSH pulse frequency was not affected under the same experimental conditions. At the same dose of tamoxifen, testicular expression of both alpha and betaB subunits of inhibin/activin was upregulated. The study demonstrated that reduced circulating LH levels were due to a decrease in hypothalamic LHRH concentration and in LH pulsatility following tamoxifen treatment. The lack of effect on circulating FSH under the same experimental conditions was likely due to its modulation by inhibin and activin.

A study of the haematologic spectrum of myelodysplastic syndrome

In 31 patients of myelodysplastic syndrome, RAEB-t was the commonest subtype (29%), and RARS, the lease common (6.4%); 19.4% were characterised as the unclassifiable (UC) group. Pallor was the dominant sign (90.3%). Low haemoglobin in RA & RARS (p<0.05), thrombocytopenia in RAEB-t (p<0.01) and high leuco/monocyte counts in CMML (p<0.001) were observed. Neutropenia occurred most frequently in RAEB & RAEB-t and circulating blasts in all cases of RAEB-t and CMML. Bicytopenia was the commonest finding (58.1%) and pancytopenia the least (16.1%). 84% of marrows were hypercellular and trilineage dysplasia was seen in 68% of patients. Megaloblastoid dyserythropoiesis was the predominant feature in all cases, dysgranulopoiesis in all cases of RAEB, RAEB-t and CMML, and micromegokaryocytes in all cases of RARS, RAEB & CMML were seen. RAEB-t and RAEB (33.3% each) were the predominant groups which progressed to leukemia, FAB AML-M2, being the commonest type (60%).

Pancreatic cancer. Cytologic study of peritoneal washings

OBJECTIVE

To delineate the frequency and significance of free cancer cells in the peritoneal cavity of patients with pancreatic exocrine adenocarcinoma.

DESIGN

Randomly selected patients with pancreatic cancer had peritoneal washings performed at the beginning and conclusion of laparotomy. Results of cytologic studies were correlated with the clinical findings, size, spread, and resectability of the tumor and with the survival time of the patient. All patients were followed up until the present or until their deaths.

SETTING

Tertiary care, referral hospitals in Toledo, Ohio.

PATIENTS

Only patients with biopsy-proven adenocarcinoma of the pancreas were included. Thirty-six patients, yielding 62 specimens for cytologic study, were included.

INTERVENTION

Peritoneal washings were performed at the beginning and completion of laparotomy. Each washing was evaluated independently by two skilled cytologists. If present, ascites was quantitated and studied cytologically. Biopsy specimens were obtained in each patient at the time of the study.

MAIN OUTCOME MEASUREMENTS

Presence or absence of malignant cells in peritoneal fluid; maximal diameter, grade, and spread of cancer; presence and volume of ascitic fluid; resectability of cancer; and length of survival of the patient.

RESULT

Of the 36 patients studied, three had positive cytologic findings. All three had peritoneal carcinomatosis. Of 11 patients with ascites, only one had positive cytologic findings.

CONCLUSIONS

Results of cytologic studies of peritoneal washings or of ascitic fluid are seldom positive with pancreatic exocrine carcinoma. When positive, they denote a very grave prognosis.

Case report 802: Ewing's tumor of the proximal phalanx of the third finger with radiographic progression documented over a 6-year-period

In summary, a case of Ewing's tumor of the proximal phalanx of the long finger of the right hand was presented. This case is unusual because of its location as well as the unusually long clinical course before the correct diagnosis was made. The major point to be stressed is that Ewing's tumor may initially present with a "benign" appearance mimicking a bone cyst [17] or angioma [12]. Timely follow-up of so-called benign lesions, particularly when recurrent swelling occurs, would help in avoiding prolonged delays in diagnosis.

Primary congenital rhabdoid tumor of the brain with neoplastic hydranencephaly

A child with shunted hydranencephaly and presumed ventriculitis was found to have a primary congenital rhabdoid tumor (RT) of the brain. The child died and a complete autopsy was carried out. The cerebral hemispheres were replaced by a single thin-walled cavity studed with tumor nodules and filled with thick, viscous fluid. The posterior fossa and visceral organs were free of tumor. This case is unique because the rhabdoid tumor was primary to the brain, it was congenital, and it massively replaced the cerebral hemispheres, causing hydranencephaly. Only three other cases of primary RT of the brain with complete autopsy examination have been reported. Cases of congenital rhabdoid tumors are not known in the literature. Hydranencephaly with a highly proteinaceous fluid should alert the physician to the possibility of a neoplasm. When the fluid in presumed ventriculitis is sterile, cerebral biopsy should be considered.

Group A variants defined with a monoclonal anti-A reagent

Unusual serologic findings mimicking the B(A) phenomenon were noted in two group B blood donors using blended murine monoclonal anti-A reagent. With additional studies, including a binding experiment using affinity chromatography, the presence of aberrant group A activities was confirmed. This observation suggests that persons with atypical B(A) phenotype may warrant further investigation to delineate the ABO blood group.