Perception without preconception: comparison between the human and machine learner in recognition of tissues from histological sections.

Deep neural networks (DNNs) have shown success in image classification, with high accuracy in recognition of everyday objects. Performance of DNNs has traditionally been measured assuming human accuracy is perfect. In specific problem domains, however, human accuracy is less than perfect and a comparison between humans and machine learning (ML) models can be performed. In recognising everyday objects, humans have the advantage of a lifetime of experience, whereas DNN models are trained only with a limited image dataset. We have tried to compare performance of human learners and two DNN models on an image dataset which is novel to both, i.e. histological images. We thus aim to eliminate the advantage of prior experience that humans have over DNN models in image classification. Ten classes of tissues were randomly selected from the undergraduate first year histology curriculum of a Medical School in North India. Two machine learning (ML) models were developed based on the VGG16 (VML) and Inception V2 (IML) DNNs, using transfer learning, to produce a 10-class classifier. One thousand (1000) images belonging to the ten classes (i.e. 100 images from each class) were split into training (700) and validation (300) sets. After training, the VML and IML model achieved 85.67 and 89% accuracy on the validation set, respectively. The training set was also circulated to medical students (MS) of the college for a week. An online quiz, consisting of a random selection of 100 images from the validation set, was conducted on students (after obtaining informed consent) who volunteered for the study. 66 students participated in the quiz, providing 6557 responses. In addition, we prepared a set of 10 images which belonged to different classes of tissue, not present in training set (i.e. out of training scope or OTS images). A second quiz was conducted on medical students with OTS images, and the ML models were also run on these OTS images. The overall accuracy of MS in the first quiz was 55.14%. The two ML models were also run on the first quiz questionnaire, producing accuracy between 91 and 93%. The ML models scored more than 80% of medical students. Analysis of confusion matrices of both ML models and all medical students showed dissimilar error profiles. However, when comparing the subset of students who achieved similar accuracy as the ML models, the error profile was also similar. Recognition of 'stomach' proved difficult for both humans and ML models. In 04 images in the first quiz set, both VML model and medical students produced highly equivocal responses. Within these images, a pattern of bias was uncovered-the tendency of medical students to misclassify 'liver' tissue. The 'stomach' class proved most difficult for both MS and VML, producing 34.84% of all errors of MS, and 41.17% of all errors of VML model; however, the IML model committed most errors in recognising the 'skin' class (27.5% of all errors). Analysis of the convolution layers of the DNN outlined features in the original image which might have led to misclassification by the VML model. In OTS images, however, the medical students produced better overall score than both ML models, i.e. they successfully recognised patterns of similarity between tissues and could generalise their training to a novel dataset. Our findings suggest that within the scope of training, ML models perform better than 80% medical students with a distinct error profile. However, students who have reached accuracy close to the ML models, tend to replicate the error profile as that of the ML models. This suggests a degree of similarity between how machines and humans extract features from an image. If asked to recognise images outside the scope of training, humans perform better at recognising patterns and likeness between tissues. This suggests that 'training' is not the same as 'learning', and humans can extend their pattern-based learning to different domains outside of the training set.

Correlation between placental histopathology and perinatal outcome in COVID-19.

Objectives: An alarming rate of adverse perinatal outcomes as well as maternal deaths has been reported worldwide during this pandemic. It would be prudent to start thinking on the lines of acute or chronic intrauterine fetal hypoxia due to placental microvascular pathology or villitis caused by severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) infection. Autopsy studies of deceased patients with severe COVID-19 have revealed the presence of diffuse pulmonary alveolar damage, thrombosis, and microvascular injuries. It is expected that similar pathological features such as microvascular injuries could be found in the placenta of infected pregnant women. Materials and Methods: Placentas of singleton pregnancies from 42 SARS-CoV-2 positive mothers delivered at term were submitted for histopathological examination. Those with multifetal gestation, hypertensive disorder, fetal growth restriction, structural or chromosomal anomalies in the fetus, thrombophilia, prolonged prelabor rupture of membranes, and placenta accreta spectrum were excluded from the study. Histopathological examination was done by two pathologists independently and only those results concurred by both were reported. Histopathological features and corresponding neonatal outcome were analyzed. Results: Reports of 42 placentas from patients with SARS-CoV-2, delivered at term (37-40 weeks) were analyzed in our study. Features of maternal vascular malperfusions (MVM) were present in 45% (n = 19) cases. Features of fetal vascular malperfusions (FVM) were present in 23.8% (n = 10) cases. There were 47.6% (n = 20) cases showing at least one feature of acute inflammatory pathology (AIP) and 42.8% (n = 18) showing features of chronic inflammatory pathology (CIP). Neonatal respiratory distress syndrome was found in 19% (n = 8) of the neonates. Correspondingly, nearly all placentas (n = 7) of these neonates showed features of MVM, FVM, AIP and CIP. There was no maternal or neonatal mortality in our study group. Conclusion: The main findings of our study include maternal as well as fetal vascular malperfusions and placental inflammatory pathology. These findings provide an outline for better understanding of etiological factors and pathogenesis of adverse perinatal outcomes in SARS-CoV-2 infection.

Performance characteristics of an artificial intelligence based on convolutional neural network for screening conventional Papanicolaou-stained cervical smears.

BACKGROUND: The conventional Papanicolaou-stained cervical smear is the most common screening test for cervical cancer. The sensitivity of the test in detecting abnormal cells is 67-75% in various studies. Owing to the volume of smears at cancer screening centres, significant man-hours are expended in the test. We have developed a software program for identification of foci of abnormal cells from conventional smears. We have chosen the convolutional neural network (CNN) model for its efficacy in image classification. METHODS: A total of 1838 microphotographs from cervical smears, containing 1301 'normal' foci and 537 'abnormal' foci were included in the study. The data set was split into training, testing and validation sets. A CNN was developed in the Python programming language. The CNN was trained with the training and testing set. At the end of training, 94.64% accuracy was achieved in the testing set. The CNN was then run on the validation set (441 images). RESULTS: The CNN showed 94.28% sensitivity, 96.01% specificity, 91.66% positive predictive value and 97.30% negative predictive value. The CNN could recognise normal squamous cells, overlapping cells, neutrophils and debris and classify the focus appropriately. False positives were reported when the CNN failed to recognise overlapping cells (2.7% microphotographs). It could correctly label cell clusters with high nuclear cytoplasmic ratio and hyperchromasia. In 1.8% of microphotographs, a false negative was reported. CONCLUSION: The CNN showed 95.46% diagnostic accuracy, suggesting potential use in screening.

A study of association between cervical cytology and period of co-habitation with husbands in the wives of serving soldiers.

BACKGROUND: The wives of serving soldiers constitute a special demographic cohort, as they spend variable amount of time in co-habitation with their husbands. The present study aims to find if any independent association, adjusted for age, exists between the time spent living together with the husband and findings on cervical smear. METHODS: Over a period of one year, cervical smears were taken from subjects at two different hospitals. The approximate length of co-habitation with husbands in the last 10 years was elicited through a questionnaire. RESULTS: Among 262 subjects, majority of smears showed normal findings (82.06%). 17.93% of the smears displayed abnormality, the maximum proportion of cases with abnormal findings were seen in the group who have spent 61-90 months, in the last ten years. Age adjusted chi square statistics failed to show statistically significant association between period of co-habitation and abnormal Pap smear. Odds' ratio (OR) for each age stratum varied from each other and was also different from the overall (crude) OR, suggesting that age is an effect modifier. Variation in epithelial cytology did not appear to be an effect of duration of cohabitation but was because of the increasing age. CONCLUSIONS: Cervical cytology does not show association with length of cohabitation with husbands in this study. However, age is seen to be an effect modifier.

Performance of A Convolutional Neural Network in Screening Liquid Based Cervical Cytology Smears.

CONTEXT: Cervical cancer is the second most common cancer in women. The liquid based cervical cytology (LBCC) is a useful tool of choice for screening cervical cancer. AIMS: To train a convolutional neural network (CNN) to identify abnormal foci from LBCC smears. SETTINGS AND DESIGN: We have chosen retrospective study design from archived smears of patients undergoing screening from cervical cancer by LBCC smears. MATERIALS AND METHODS: 2816 images, each of 256 × 256 pixels, were prepared from microphotographs of these LBCC smears, which included 816 "abnormal" foci (low grade or high grade squamous intraepithelial lesion) and 2000 'normal' foci (benign epithelial cells and reactive changes). The images were split into three sets, Training, Testing, and Evaluation. A convolutional neural network (CNN) was developed with the python programming language. The CNN was trained with the Training dataset; performance was assayed concurrently with the Testing dataset. Two CNN models were developed, after 20 and 10 epochs of training, respectively. The models were then run on the Evaluation dataset. STATISTICAL ANALYSIS USED: A contingency table was prepared from the original image labels and the labels predicted by the CNN. RESULTS: Combined assessment of both models yielded a sensitivity of 95.63% in detecting abnormal foci, with 79.85% specificity. The negative predictive value was high (99.19%), suggesting potential utility in screening. False positives due to overlapping cells, neutrophils, and debris was the principal difficulty met during evaluation. CONCLUSIONS: The CNN shows promise as a screening tool; however, for its use in confirmatory diagnosis, further training with a more diverse dataset will be required.

Pilot Study of an Open-source Image Analysis Software for Automated Screening of Conventional Cervical Smears.

INTRODUCTION: The Pap stained cervical smear is a screening tool for cervical cancer. Commercial systems are used for automated screening of liquid based cervical smears. However, there is no image analysis software used for conventional cervical smears. The aim of this study was to develop and test the diagnostic accuracy of a software for analysis of conventional smears. MATERIALS AND METHODS: The software was developed using Python programming language and open source libraries. It was standardized with images from Bethesda Interobserver Reproducibility Project. One hundred and thirty images from smears which were reported Negative for Intraepithelial Lesion or Malignancy (NILM), and 45 images where some abnormality has been reported, were collected from the archives of the hospital. The software was then tested on the images. RESULTS: The software was able to segregate images based on overall nuclear: cytoplasmic ratio, coefficient of variation (CV) in nuclear size, nuclear membrane irregularity, and clustering. 68.88% of abnormal images were flagged by the software, as well as 19.23% of NILM images. The major difficulties faced were segmentation of overlapping cell clusters and separation of neutrophils. CONCLUSION: The software shows potential as a screening tool for conventional cervical smears; however, further refinement in technique is required.

Artificial Intelligence in Cytopathology: A Neural Network to Identify Papillary Carcinoma on Thyroid Fine-Needle Aspiration Cytology Smears.

INTRODUCTION: Fine-needle aspiration cytology (FNAC) for identification of papillary carcinoma thyroid is a moderately sensitive and specific modality. The present machine learning tools can correctly classify images into broad categories. Training software for recognition of papillary thyroid carcinoma on FNAC smears will be a decisive step toward automation of cytopathology. AIM: The aim of this study is to develop an artificial neural network (ANN) for the purpose of distinguishing papillary carcinoma thyroid and nonpapillary carcinoma thyroid on microphotographs from thyroid FNAC smears. SUBJECTS AND METHODS: An ANN was developed in the Python programming language. In the training phase, 186 microphotographs from Romanowsky/Pap-stained smears of papillary carcinoma and 184 microphotographs from smears of other thyroid lesions (at ×10 and ×40 magnification) were used for training the ANN. After completion of training, performance was evaluated with a set of 174 microphotographs (66 - nonpapillary carcinoma and 21 - papillary carcinoma, each photographed at two magnifications ×10 and ×40). RESULTS: The performance characteristics and limitations of the neural network were assessed, assuming FNAC diagnosis as gold standard. Combined results from two magnifications showed good sensitivity (90.48%), moderate specificity (83.33%), and a very high negative predictive value (96.49%) and 85.06% diagnostic accuracy. However, vague papillary formations by benign follicular cells identified wrongly as papillary carcinoma remain a drawback. CONCLUSION: With further training with a diverse dataset and in conjunction with automated microscopy, the ANN has the potential to develop into an accurate image classifier for thyroid FNACs.

A case of mature cystic teratoma arising from the fourth ventricle.

Intracranial teratomas represent a rare lesion accounting for 0.1%-0.7% of all intracranial tumors. Those in the fourth ventricle have rarely been reported. The present case is that of a 28-year-old man with occipital headache for two months. MRI examination revealed a well-defined extra-axial cystic lesion in posterior fossa in the midline herniating through the foramen magnum. Pre operatively, the mass was seen to be occupying the whole of the posterior fossa and arising from the roof of the fourth ventricle. On gross examination, the lesion had both solid and cystic components. Histopathological examination showed multiple cystic areas lined by brain tissue admixed with islands of cartilage and salivary gland elements and intestinal type glands. A diagnosis of mature cystic teratoma was made.