IOT-Based Medical Informatics Farming System with Predictive Data Analytics Using Supervised Machine Learning Algorithms.

In the farming industry, the Internet of Things (IoT) is crucial for boosting utility. Innovative agriculture practices and medical informatics have the potential to increase crop yield while using the same amount of input. Individuals can benefit from the Internet of Things in various ways. The intelligent farms require the creation of an IoT-based infrastructure based on sensors, actuators, embedded systems, and a network connection. The agriculture sector will gain new advantages from machine learning and IoT data analytics in terms of improving crop output quantity and quality to fulfill rising food demand. This paper described an intelligent medical informatics farming system with predictive data analytics on sensing parameters, utilizing a supervised machine learning approach in an intelligent agricultural system. The four essential components of the proposed approach are the cloud layer, fog layer, edge layer, and sensor layer. The primary goal is to enhance production and provide organic farming by adjusting farming conditions as per plant needs that are considered in experimentation. The use of machine learning on acquired sensor data from a prototype embedded model is investigated for regulating the actuators in the system. Then, an analytics and decision-making system was built at the fog layer, employing two supervised machine learning approaches including classification and regression algorithms using a support vector machine (SVM) and artificial neural network (ANN) for effective computation over the cloud layer. The experimental results are evaluated and analyzed in MATLAB software, and it is found that the classification accuracy using SVM is much better as compared to ANN and other state of art methods.

In vivo antinuclear antibodies of the skin: A rare phenomenon on direct immunofluorescence.

Detection of immunoglobulins in epidermal cell nuclei or in vivo antinuclear antibodies on direct immunoflorescence microscopy of skin biopsies is an easily detectable immunopathologic feature. It is an unusual, but not totally rare, occurrence in systemic connective tissue disorders. This positive epidermal nuclear reaction is found to be commonly associated with immunoglobulin G.

Fine needle aspiration cytology of Sertoli-Leydig cell tumors of ovary masquerading as dysgerminoma.

Herein, we described a case of a 29-year-old female with a large ovarian mass. Fine needle aspiration cytology (FNAC) of the mass showed discrete round to oval cells in a fatty vacuolated background. FNAC diagnosis of dysgerminoma was suggested. The histology of the tumors showed features of poorly differentiated Sertoli-Leydig cell tumors. We discussed the diagnostic pitfalls of this case on FNAC.

Non-Hodgkin's lymphoma presenting as breast masses: a series of 10 cases diagnosed on FNAC.

Primary breast lymphoma (PBL) is a rare disease, which comprises 0.04-0.53% of all primary malignant tumors of the breast. The most frequent histological subtype is diffuse large B-cell type (DLBCL) (40-70%). Differentiation of PBLs from other breast tumors such as poorly differentiated carcinomas and lobular carcinoma may at times be difficult on cytomorphology alone. An audit of breast lymphomas diagnosed on fine needle aspiration cytology (FNAC) over a period of 9 years (2001-2009) was performed. Ten cases were retrieved and the cytomorphology was reviewed along with immunochemistry (IHC), flow cytometry as well as histopathology, wherever available. The age of patients ranged from 17 to 83 years. Eight cases were diagnosed as non-Hodgkin's lymphoma, high-grade on FNAC. Histopathology was available in four of these cases and cell block was available in one case. Lymphoid cells were positive for leukocyte-common antigen (LCA) and CD20 and negative for CD3 in these cases. The same was confirmed by flow cytometry on aspirated material in one case. A diagnosis of DLBCL was offered in these five cases. One case was a low-grade NHL and another case was a young male, a known case of acute leukemia and had leukemic infiltration in the breast lump. We wish to emphasize the potential importance of FNAC in breast lymphoma and the same can be helpful to avoid unnecessary surgery in these cases. The differential diagnostic entities have been discussed. IHC and flow cytometry can be performed on the aspirated material and provide valuable information.

Fine-needle aspiration in the evaluation of thyroid lesions in children.

The role of fine-needle aspiration (FNA) cytology in the evaluation of thyroid lesions in not as well established in children when compared with adults. Hence we aimed to ascertain the utility and limitations of FNA in childhood thyroid lesions. This was a retrospective analysis of all thyroid FNA performed in children less than 14 years of age over a 4-year period (2005-2009). Histopathological follow-up was available in six cases. A total of 77 cases were included in the analysis. The most common cytological diagnosis was lymphocytic thyroiditis (49.3%), followed by colloid goiter (18.2%), hyperplasia (10.4%), and benign aspirate (7.8%); malignancy was identified in six cases (7.8%). Of these six cases, three were papillary thyroid carcinoma. There was one false-positive case reported as a Hurthle-cell neoplasm, which on histology showed Hashimoto's thyroiditis. One case each of rhabdomyosarcoma and spindle epithelial tumor with thymus like differentiation was wrongly diagnosed as thyroid neoplasm, NOS, and medullary carcinoma (spindle variant), respectively. The overall diagnostic accuracy was 98.6% with 100% sensitivity, 98.6% specificity, 80% positive predictive value, and 100% negative predictive value. FNA is extremely valuable in the initial evaluation of thyroid swelling in children. Rare neoplasms masquerading as thyroid nodules in children can pose difficulties in diagnosis; however, papillary carcinoma is easily recognized. In lymphocytic thyroiditis, it provides a tissue diagnosis, thereby avoiding more invasive procedure for merely diagnostic purposes.

Intranuclear peudoinclusions: morphology, pathogenesis, and significance.

Intranuclear pseudoinclusions represent invaginations of the cytoplasm into the nucleus. The nuclear pseudoinclusions are generally easier to appreciate in cytological preparations than frozen sections or histopathology thus forming an important cytological feature. In the present article we have reviewed the morphology, pathogenesis, differential diagnosis, and clinical significance of intranuclear cytoplasmic pseudoinclusion.

Malignant biphasic peritoneal mesothelioma in a child: fine-needle aspiration cytology, histopathology, and immunohistochemical features along with review of literature.

Peritoneal mesotheliomas in children are of rare occurrance. We herein report the clinical, radiological, and pathological findings of a rare case of malignant peritoneal mesothelioma occurring in nine-year-old female child. The child presented with abdominal distension and awareness of a painless mass in the abdomen which on radiology appeared as a large heterogeneous pelvic mass with peritoneal deposits at multiple sites. To the best of our knowledge, this is the first case of a peritoneal malignant mesothelioma on which fine needle aspiration (FNA) was performed as first line investigation of the primary tumor. The cytological features, major differential diagnoses, and the pitfalls therein are discussed. Histopathology revealed biphasic pattern of mesothelioma which is again a rare pattern. Immunochemistry was carried out on the cell block made from the FNA as well as the biopsy specimen essentially showed the same features. There was positivity for vimentin, EMA, and cytokeratin 5/6 while WT1, calretinin, and CEA were negative; however, D2-40 showed diffuse membranous positivity in the epithelial areas and cytoplasmic positivity in the spindle areas confirming a mesothelioma. We emphasize the use of immunochemistry on cell block material for a confident diagnosis of mesothelioma in such cases.

Fractal and lacunar dimension of the chorionic villi of molar pregnancy.

OBJECTIVE: To explore the role of fractal geometry to differentiate molar pregnancies from normal chorionic villi. STUDY DESIGN: We selected at random 15 cases each of first-trimester normal chorionic villi, complete hydatidiform mole, and partial mole. Fractal dimension (FD) and lacunar dimension (LD) were measured in each case using the box counting method. FD and LD were compared in the three groups. RESULTS: Mean FD was highest in the normal chorionic villi (1.7520), followed by partial mole (1.6696) and complete mole (1.6438). Analysis of variance (ANOVA) showed a significant difference of FD in normal villi vs. complete (p < 0.001) and partial mole (p < 0.001). There was no significant difference of FD between complete and the partial mole. The mean LD of normal, partial, and complete molar villi was 0.5067 +/- 0.6944, 0.6063 +/- 0.09670, and 0.5551 +/- 0.11277, respectively. The mean LD was significantly increased between partial mole and normal villi (ANOVA, p < 0.006). However, there was no significant difference of LD between the partial and complete mole and between complete mole and normal villi. CONCLUSION: The measurement of FD and LD along with the histopathology of the lesions may be helpful to distinguish molar and nonmolar villi.

Epithelioid cell granulomas in urine cytology smears: Same cause, different implications.

Tuberculosis of the urinary tract is usually secondary to tuberculosis of the kidney. Multinucleated giant cell histiocytes, often with peripheral nuclei (Langhans' cells), may be identified. Acid-fast bacilli on smear or positive urine cultures confirm the diagnosis. Similar findings can also be seen in patients treated with Bacillus Calmette Guérin (BCG) for transitional cell carcinoma or after bladder surgery. Here, we present two cases showing epithelioid cell granulomas and multinucleated giant cells on urine cytology, and discuss the differentiating features on cytomorphology and their therapeutic implications.

Micronucleus in atypical urothelial cells.

The objective of this study is to evaluate the role of micronuclei (MN) in identifying the atypical urothelial cells urinary cytology. This is a retrospective study of a total of 30 cases of urine samples. Of these, 15 cases were selected where the initial diagnosis was atypical urothelial cell and later turned out as malignant cells on follow-up histopathology. The rest of the 15 cases were reported as normal urothelial cells. The number of MN in 1,000 urothelial cells was counted in all the cases and was expressed as percentage. All the cases diagnosed as atypical urothelial cells show the presence of MN. The mean number of MN was 2.53 ± 0.99%. No MN was noted in any of the normal urothelial cells in control cases. The Student's t-test of MN scoring was highly significant (P < 0.000) in test (atypical urothelial cells) versus control group (benign urothelial cells). MN is an important biomarker of malignancy. Micronuclear score in atypical urothelial cells in urine cytology smears may help in proper diagnosis of the case.