Search

Purpose The purpose of the proposed model is to assist the e-business to predict the churned users using machine learning. This paper aims to monitor the customer behavior and to perform decision-making accordingly. Design/methodology/approach The proposed model uses the 2-D convolutional neural network (CNN; a technique of deep learning). The proposed model is a layered architecture that comprises two different phases that are data load and preprocessing layer and 2-D CNN layer. In addition, the Apache Spark parallel and distributed framework is used to process the data in a parallel environment. Training data is captured from Kaggle by using Telco Customer Churn. Findings The proposed model is accurate and has an accuracy score of 0.963 out of 1. In addition, the training and validation loss is extremely less, which is 0.004. The confusion matric results show the true-positive values are 95% and the true-negative values are 94%. However, the false-negative is only 5% and the false-positive is only 6%, which is effective. Originality/value This paper highlights an inclusive description of preprocessing required for the CNN model. The data set is addressed more carefully for the successful customer churn prediction.

External auditory canal, like other epithelialized surfaces is predisposed to malignancies such as Squamous cell carcinomas (SCC), melanomas and adenocarcinomas. In this background, malignancies like adenoid cystic carcinoma (ACC) are a rare occurrence and need to be thoroughly evaluated both locally and for distant extension. The malignancy needs to be addressed with an aggressive approach surgically with adequate marginal clearance. The role of radiation is debatable considering the outcomes in the limited data. Here we present a case of a 35-year-old female who presented with otalgia and otorrhea associated a mass in the external auditory canal. After biopsy and imaging, a diagnosis of adenoid cystic carcinoma was made, and the patient was managed accordingly. Keywords: Adenoid cystic carcinoma, External auditory canal, Neck dissection Continuous...

Dental extraction of abscessed or infected teeth before cardiac operation is frequently advocated to lower the risk of infections especially infective endocarditis. The scientific evidence that supports dental procedures before cardio-valvular surgery is however limited. The aim of the present paper was to explore whether there are any protocols for patients needing dental treatment before cardiac operations. Moreover, we have discussed the real life challenges encountered in the management of such patients. Continuous....

Background Retiform Hemangioendothelioma (RH) is an extremely rare vascular tumor of intermediate biological behavior, which is prone to local recurrence but rarely shows metastasis to distant sites. It may harbor areas resembling Dabska tumor in some cases and angiosarcoma, which in its well differentiated form may exhibit similar pathological appearance in some areas, making it problematic to rule out a possibility of a malignant diagnosis on a core biopsy. Therefore, complete surgical resection with negative margins is essential for accurate diagnosis and local control. Results In our series, two of the three Pakistani cases were in females, with an age range between 18 and 50 years. Our first patient presented with symptoms of cardiac compromise and pulmonary hypertension. Her computed tomography scan showed multiple tumor masses within the mediastinum. The second patient presented with an ulcerated lesion on his scalp, at right temple. The third patient presented with a hard growth on her left 4th toe which was amputated. Histologically, all cases exhibited retiform arborizing vascular spaces lined by bland endothelial cells with hobnail nuclei, characteristic of retiform hemangioendothelioma. Immunohistochemical markers CD31, CD34 and ERG confirmed the vascular nature of the tumor. The first and the second patient are alive and healthy at 4 and 7 months follow up respectively, while the third patient is lost to follow up. Conclusion Owing to the rate of local recurrence, RH should always be considered in the differential diagnosis of vascular tumors showing arborizing blood vessels, as it may have an atypical presentation and it should be carefully differentiated from Dabska tumor and an angiosarcoma.

Highlights • Diagnosis and management of giant juvenile fibroadenoma can be challenging. • Maintaining symmetry after complete excision is tougher in developing breasts. • Periareolar approach for subareolar fibroadenomas provides good cosmetic results. • Benign tumor to breast size ratio up to 70% provides excellent cosmetic outcome., Introduction Fibroadenoma is the most common benign lesion of breast in young women, characterized by an aberrant proliferation of both epithelial and mesenchymal elements. It is termed giant fibroadenoma when it is larger than 5 cm or weighs more than 500 g with an incidence of 0.5–2% of all fibroadenomas. Presentation of case In this report, we discuss a case of a 13-year-old Pakistani girl who presented with a giant juvenile fibroadenoma in left breast and was treated by a subareolar lump excision through a periareolar incision with excellent cosmetic outcome. To the best of our literature search, this is the first case of giant juvenile fibroadenoma in an adolescent being reported from Pakistan. Discussion Surgical management of giant juvenile fibroadenoma in immature breast is challenging as it may either result in asymmetric defect or damage to developing breast tissue resulting in long term poor outcomes. Surgical decision should be carefully undertaken and reported for future reference in such cases. Conclusion The diagnosis and management of giant juvenile fibroadenoma can be challenging because these tumors clinically and histologically mimic phyllodes tumor due to their rapid growth and large size. Excision through a periareolar approach for fibroadenomas located in subareolar region provides good cosmetic results in these patients with minimal scar visibility.

We observed an unprecedented and consistent low performance of senior residents as compared with juniors in monthly examinations. This compelled us to evaluate systematically and compare the scores of senior residents with their junior colleagues. This retrospective observational study was conducted in April 2020 among internal medicine residents. Residents in first and second year of their training were labelled as junior residents. Residents in third or fourth year of their training were labelled as senior residents. Comparison of mean scores of each resident level was done separately both for monthly formative multiple-choice questions tests, and summative yearly end of term examinations. We discussed the possible reasons as well. There were 67 residents in year 2018 and 69 in 2019. There is no significant difference between scores of monthly examinations of years 2018 and 2019 among residents of each level. Rather, in March and December 2018, junior residents perform better than senior residents with p values of 0.01 and 0.04, respectively. In February and September 2019, senior residents performed better than junior residents with p value of 0.02. Similarly, there is no significant difference in scores of residents of each level in end-of-term examinations of years 2018 and 2019 with p values 0.18 and 0.25, respectively. Performance of senior residents in our residency programme in in-service examinations is relatively low as compared with their junior colleagues. There is a need to evaluate reasons for this relatively low performance of senior residents.

The energy consumption of an enterprise energy management system (EMS) is a complex process with nonlinearity, time-variance, larger delay, greater inertia and other dynamic characteristics, resulting in the failure of a single-item prediction model to achieve satisfactory prediction results. In this paper, a combination prediction method, based on BP neural network, was proposed to predict the energy consumption of an enterprise EMS for improving the prediction accuracy. The energy consumption of enterprise energy management system (EMS) was predicted and analyzed using gray combination models, i.e., GM (1.1) and pGM (1.1), gray Markov chain, and BP neural network prediction model. These single-item models and their prediction processes were constructed and successfully applied to predict the energy consumption of iron and steel enterprises. The data pertaining to energy consumption of these enterprises from January to December 2018 and January to March 2019 were used for predicting the simulation and testing, respectively. The results showed that the prediction results of our approach has an average relative error of 3.327% and 1.298% respectively, which are extremely lower than the existing approaches for improving the prediction accuracy.

Objectives: Congenital heart defects (CHDs) affect more than 40,000 children annually in Pakistan. Approximately 80% of patients require at least one surgical intervention to achieve a complete or palliative cardiac repair. The Glenn shunt, a palliative procedure is established between superior vena cava (SVC) and the right pulmonary artery to provide an anastomosis offering minimal risk to patients with univentricular heart disease. The aim of this study was to assess the clinical outcomes of the Glenn shunt procedure in patients with complex congenital heart diseases in a developing country like Pakistan. Materials and Methods: A retrospective chart review was conducted on patients who underwent a bidirectional Glenn shunt procedure from July 2006 to June 2017. Data were collected on a structured questionnaire and analyses performed on SPSS version 22. Frequencies and percentages were computed for categorical variables while mean and standard deviation for continuous variables where appropriate. Results: A total of 79 patients underwent the Glenn shunt procedures. The median age was 1.9 years and 54.5% were male. Tricuspid atresia was the primary diagnosis in 30.4% of the patients. Common morbidities included arrhythmias (6.3%), pleural effusion (8.9%), wound infection (3.8%), pneumonia (2.5%), and seizures (3.8%); reopening was required in 2.5% of the patients and 8.8% were readmitted within 30 days of index operation. There were three (3.8%) deaths in total. Conclusions: Bidirectional Glenn shunt procedure can be performed safely in patients with ideal characteristics as the first stage palliation and has favorable results with acceptable rate of complications.

First described in 1934, eccrine spiradenoma (ES) is a rare, benign adnexal tumour arising from eccrine sweat glands. It commonly presents as a slow-growing nodule on the upper trunk, and head and neck region, mostly in the age bracket of 15-35 years, with no gender preference. While no established guidelines exist for optimal management of malignant ES, some therapies have been studied. The diagnosis of this entity is extremely important as it can harbour a malignant component with disastrous outcomes which may be missed due to its strong resemblance to benign lesions, such as a papilloma. Here, we present the case of a 35-year-old lady who presented with a papilloma-like growth on the upper medial aspect of the thigh which was diagnosed as eccrine spiradenoma upon excision. Keywords: Eccrine Spiradenoma, adnexal tumour, dermal lesion, case report. Continuous...

Background Cognitive apprenticeship model (CAM) is an instructional model for situated learning. There is limited data available on application of the CAM in clinical settings. The aim of the study was to identify learning strategies using CAM, which in the opinion of learners are effective in ward rounds. Methods Participants were residents and medical students who rotated through internal medicine at Aga Khan University Hospital, Karachi. We sought learners’ opinion on a structured questionnaire based on four principal dimensions of cognitive apprenticeship. A previously determined set of 10 defined competencies were compared with CAM’s six teaching/learning (T/L) methods (modelling, coaching, scaffolding, articulation, reflection and exploration) as well. Mean and SD were calculated. Mann-Whitney test was used to compare scores. Results Of 195 participants, there were 100 men (51.3%) and 95 women (48.7%). Perceived learning for six T/L methods, ranged from 3.7 to 3.9 (max=5). Coaching and scaffolding had the highest scores. Statistically significant difference between the student and resident groups was noted. Medical students perceived coaching and scaffolding (4.1 and 4.05) and residents rated coaching, articulation and exploration as most effective (3.9 and 3.8). Majority (82.1%) reported a positive learning environment in wards. Conclusions CAM enabled to identify two T/L methods (scaffolding and coaching) that are important for learning in ward round. Limited differences in perceived effectiveness of the T/L methods indicate that variety can be used to sustain interest in learners. Positive learning environment, team diversity and tasks of increasing complexity contribute to learning.

Data Warehouse (DW) applications provide past detail for judgment process for the companies. It is acknowledged that these systems depend on Multidimensional (MD) modelling different from traditional database modelling. MD modelling keeps data in the form of facts and dimensions. Some proposals have been presented to achieve the modelling of these systems, but none of them covers the MD modelling completely. There is no any approach which considers all the major components of MD systems. Some proposals provide their proprietary visual notations, which force the architects to gain knowledge of new precise model. This paper describes a framework which is in the form of an extension to Unified Modelling Language (UML). UML is worldwide known to design a variety of perspectives of software systems. Therefore, any method using the UML reduces the endeavour of designers in understanding the novel notations. Another exceptional characteristic of the UML is that it can be extended to bring in novel elements for different domains. In addition, the proposed UML profile focuses on the accurate representations of the properties of the MD systems based on domain specific information. The proposed framework is validated using a specific case study. Moreover, an evaluation and comparative analysis of the proposed framework is also provided to show the efficiency of the proposed work

Primary anaplastic large cell lymphoma of the trachea is a very rare tumor. Diagnosis is often missed for a long time because the presentation often resembles that of obstructive airway disease. A 24-year-old non-smoker male presented at an outpatient clinic complaining of gradual development of stridor and shortness of breath over a recent period of four weeks. Imaging workup revealed a large lobulated, soft tissue, mildly enhancing, eccentrically placed intraluminal proximal tracheal mass, which showed extension outside of the tracheal lumen in the peritracheal fat. The patient underwent surgery. A biopsy of the resected mass reported anaplastic large cell lymphoma. Post-operation, the patient underwent chemotherapy consisting of four cycles of CHOP (cyclophosphamide, doxorubicin, vincristine and prednisolone). Follow-up imaging showed non-visualization of the mass. Primary lymphoma (anaplastic large cell lymphoma, which is a rare type of non-Hodgkin’s lymphoma) of the trachea is a rare tumor. Early curative resection with post-operative chemotherapy has a favorable outcome, as in our case.

Madiha Bilal Qureshi,1 Muhammad Usman Tariq,1 Jamshid Abdul-Ghafar,2 Qurratulain Chundriger,1 Nasir Ud Din1 1Department of Pathology and Laboratory Medicine, Aga Khan University Hospital, Karachi, Pakistan; 2Department of Pathology and Clinical Laboratory, French Medical Institute for Mothers and Children (FMIC), Kabul, AfghanistanCorrespondence: Jamshid Abdul-GhafarDepartment of Pathology and Clinical Laboratory, French Medical Institute for Mothers and Children (FMIC), Kabul, AfghanistanTel +93 792 827 287Email jamshid.jalal@fmic.org.afBackground: Synovial sarcoma (SS) is a high-grade spindle cell tumor that accounts for 5% to 10% of soft tissue sarcomas. The majority originate from the deep intramuscular soft tissues of extremities with common sites including knee, ankle and feet. Immunohistochemical (IHC) stain TLE1 (transducer-like enhancer of split 1) is a potent diagnostic marker for distinguishing SS from mimicking tumors.Methodology: The study was performed on 177 tumor cases, including 89 SS and 88 non-synovial sarcoma (N-SS) cases which were diagnosed at Department of Pathology and Laboratory Medicine, Aga Khan University Hospital, from July 2019 to June 2020. Hematoxylin and eosin (H&E) and IHC stained glass slides of these cases were reviewed. TLE1 expression was assessed based on the Remmele scoring system.Results: Eighty-nine cases of SS and 88 cases of N-SS were included in the study. SS cases included 42 (47.2%) monophasic subtype, 6 (6.7%) biphasic subtype and 41 (46.1%) poorly differentiated subtype. Major tumor types in N-SS cases were 27 (30.7%) Ewing sarcoma (ES), 13 (14.8%) leiomyosarcoma, 10 (11.4%) undifferentiated sarcoma (US), 8 (9.1%) fibrosarcomatous dermatofibrosarcoma protuberans and 7 (8%) malignant peripheral nerve sheath tumor cases. Mean patients’ age for SS cases was 26.14 years and for N-SS cases was 32.64 years. All 89 SS cases showed positive TLE1 expression. Out of 88 N-SS cases, 71 (80.7%) were TLE1 negative and 17 (19.3%) showed positive expression.Conclusion: This study shows that TLE1 is a very sensitive and relatively specific IHC marker for SS. TLE1 expression can be observed in other soft tissue sarcomas but diffuse strong TLE1 expression is highly specific for SS. The diagnosis should not solely rely on TLE1 expression and morphologic features but should include soft tissue specific lineage markers to avoid misdiagnosis.Keywords: synovial sarcoma, monophasic synovial sarcoma, TLE1

Background Cemento-ossifying fibroma (COF) and cementoblastoma (CB) are rare benign odontogenic tumors with a predilection for the mandible. Cemento-ossifying fibroma is a fibro-osseous lesion that originates in the tooth bearing areas of jaw and shows cementum-like tissue in a fibrotic stroma. Cementoblastoma is classically related to roots of teeth with the presence of calcified cementum-like material. To date, only a single case of concomitant unilateral COF and CB has been reported in the literature. Case presentation We present an unusual case of a 37-year-old female who presented with two discrete bilateral swellings in the right and left mandible for 10 years. The larger tumor involved the left posterior mandible with extension anteriorly to the left and right anterior mandibles, and the smaller tumor was present in right posterior mandible. Radiology revealed two distinct lesions involving both sides of mandible. Histopathological examination showed characteristic features of cemento-ossifying fibroma in sections of the larger tumor and cementoblastoma in sections of smaller tumor. Conclusion This case shows the very unique bilateral co-existence of COF and CB, the second case reported in literature to date.

As artificial intelligence's potential and pervasiveness continue to increase, its strategic importance, effects, and management must be closely examined. Societies, governments, and business organizations need to view artificial intelligence (AI) technologies and their usage from an entirely different perspective. AI is poised to have a tremendous impact on every aspect of our lives. Therefore, it must have a broader view that transcends AI's technical capabilities and perceived value, including areas of AI's impact and influence. Nicholas G. Carr's seminal paper “IT Does not Matter (Carr, 2003) explained how IT's potential and ubiquity have increased, but IT's strategic importance has declined with time. AI is poised to meet the same fate as IT. In fact, the commoditization of AI has already begun. This paper presents the arguments to demonstrate that AI is moving rapidly in this direction. It also proposes an artificial intelligence-based organizational framework to gain value-added elements for lowering the impact of AI commoditization.

The interaction among different Internet of Things (IoT) sensors and devices become massive and insecure over the Internet as we probe to smart cities. These heterogeneous devices produce an enormous amount of data that is vulnerable to various malicious threats. The generated data need to be processed and analyzed in a secure fashion to make smart decisions. The smart urban planning is becoming a reality through the mass information generated by the Internet of Things (IoT). This paper exhibits a novel architecture, SafeCity, that limelight the ecosystem of smart cities consists of cameras, sensors, and other real-world physical devices. SafeCity is a three-layer architecture, i.e., a data security layer, a data computational layer, and a decision-making layer. At the first layer, payload-based symmetric encryption is used to secure the data from intruders by exchanging only the authentic data among the physical devices. The second layer is used for the computation of secured data. Finally, the third layer extracts visions from data. The secured exchange of data is ensured by using Raspberry Pi boards while the computation of data is tested on trustworthy datasets, using the Hadoop platform. The assessments disclose that SafeCity presents precious insights into a secured smart city in the context of sensors based IoT environment.

Background Mitral valve aneurysm (MVA) is a saccular outpouching of the mitral leaflet which expands on systole and collapses during diastole. The case of MVA was first described in 1729 by Morand. It is one of the rare entities with a reported incidence of only 0.2–0.29% and no such case reported in Pakistan before. Case presentation A 51 year old female presented with dyspnea and chest pain for 3 months. Upon investigating, trans-esophageal echocardiography (TEE) revealed thickened anterior mitral valve leaflet with rolled up margins, forming an aneurysm, with severe mitral regurgitation. Subsequently, the valve was evaluated intra-operatively for repair but eventually had to be excised and then successfully replaced with a bioprosthetic valve. Conclusions TEE is an excellent technique to confirm the diagnosis of a mitral valve leaflet aneurysm, and depending upon the severity of the defect, valve repair can be attempted but replacement becomes the most suitable treatment modality, once repair is deemed impossible. We hereby report a rare case, where timely diagnosis, appropriate surgical intervention and regular post-operative follow up helped in achieving good prognosis of this rare entity.

This paper presents an in-depth literature review on the driving forces and barriers for achieving operational excellence through artificial intelligence (AI). Artificial intelligence is a technological concept spanning operational management, philosophy, humanities, statistics, mathematics, computer sciences, and social sciences. AI refers to machines mimicking human behavior in terms of cognitive functions. The evolution of new technological procedures and advancements in producing intelligence for machines creates a positive impact on decisions, operations, strategies, and management incorporated in the production process of goods and services. Businesses develop various methods and solutions to extract meaningful information, such as big data, automatic production capabilities, and systematization for business improvement. The progress in organizational competitiveness is apparent through improvements in firm’s decisions, resulting in increased operational efficiencies. Innovation with AI has enabled small businesses to reduce operating expenses and increase revenues. The focused literature review reveals the driving forces for achieving operational excellence through AI are improvement in computing abilities of machines, development of data-based AI, advancements in deep learning, cloud computing, data management, and integration of AI in operations. The barriers are mainly cultural constraints, fear of the unknown, lack of employee skills, and strategic planning for adopting AI. The current paper presents an analysis of articles focused on AI adoption in production and operations. We selected articles published between 2015 and 2020. Our study contributes to the literature reviews on operational excellence, artificial intelligence, driving forces for AI, and AI barriers in achieving operational excellence.

A 21-year-old man presented with chest pain, diaphoresis and dyspnoea. Electrocardiogram (ECG) showed inferior ST-elevation myocardial infarction. Troponin I was positive. Patient underwent left heart catheterisation, which revealed normal epicardial coronary arteries except for right coronary artery which could not be engaged. CT coronary angiogram was done, which revealed large right sinus of Valsalva aneurysm giving rise to a pinched out right coronary artery. Patient underwent composite graft replacement of aortic valve, aortic root and ascending aorta along with a saphenous vein graft to right coronary artery. He was discharged in stable condition on fifth postoperative day. Biopsy of the aneurysmal tissue was suggestive of vasculitic aetiology. There was no evidence of systemic vasculitis and Magnetic Resonance Angiography (MRA) screen was negative for Takayasu’s arteritis. Our patient was, hence, diagnosed with clinically isolated aortitis leading to ST elevation myocardial infarction due to an unruptured sinus of Valsalva aneurysm.

Background Neo-adjuvant chemotherapy (NAC) is frequently administered in breast carcinoma patients. The clinical response to NAC guides further treatment. The pathological response is not only an independent prognostic factor, but it also guides further treatment and prognosis. Objectives The aim of our study was to find the degree of concordance between clinical and pathological response assessments after NAC in Invasive lobular Carcinoma (ILC) cases by using World Health Organization (WHO) criteria and different pathological systems, respectively. We also tried to identify any useful parameter of clinical assessment that could better correlate with pathologic assessment and provide a better estimation of residual tumor. Methods This retrospective study was conducted on 26 ILC tumors diagnosed in 24 patients who were treated with NAC followed by surgical resection between January 2009 and December 2020. Medical records and microscopy glass slides were reviewed for clinical and pathological response assessments, respectively. Results The pre-treatment tumor area ranged from 1.8-255 cm2 and the mean±SD was 52.2±66.8 cm2. After NAC, complete clinical response was observed in four (15.3%) cases. The clinically assessed mean tumor area significantly reduced from 52.2±66.8 cm2 to 17.2±22.6 cm2 (p-value

Primary central nervous system lymphoma (PCNSL) is a rare type of non-Hodgkin's lymphoma. It is defined as lymphoma of the central nervous system without any systemic disease elsewhere at the time of diagnosis. Based on the phenotypical features, it is divided into two categories, B-cell and T-cell, with the latter being less common. Viruses, such as human immunodeficiency virus (HIV) and human T-lymphotropic virus type 1 (HTLV-1), are linked to the T-cell variant; however, there is a lack of evidence suggesting associating hepatitis B and D virus coinfection with it. We report a case of a 34-year-old male who presented with T-cell PCNSL and was later diagnosed with a hepatitis B and D virus coinfection.

The product development process is one of the most challenging stages of a product life cycle due to several reasons. Having the right knowledge environment during the design process may eliminate the waste of cost and time. The aim of this paper is to demonstrate a case study where designers can investigate the conflicting parameters about a product and make their decisions based on an accurate knowledge environment that is created by trade-off curves. The product in consideration is a turbofan jet engine with a requirement of noise reduction during takeoff while keeping up with high quality standards.

Background Kaposiform Hemangioendothelioma (KHE) is a rare vascular tumor of intermediate malignant potential which shows locally aggressive growth but only rarely metastasizes. It is mostly considered to be a tumor of pediatric population but its occurrence in the adults is not uncommon as once considered. Histologically, KHE can mimic other soft tissue neoplasms of different behaviors (e.g. Kaposi Sarcoma, hemangioma) and establishing the correct diagnosis is important for appropriate treatment. Herein, we describe the clinicopathological features of 8 cases of KHE which will be helpful in making their diagnosis. Methods We reviewed pathology reports, microscopy glass slides and obtained follow up information about 8 cases of KHE which were diagnosed at our institution from January 2008 till June 2020. Immunohistochemical stain for HHV8 was also performed. Results Age ranged from 7 months to 25 years. Seven patients were less than 20 years of age and one patient was 25 years old. Equal gender distribution was observed. Extremities were the most common sites of involvement, followed by head and neck, pancreas and ischiorectal region. 2 cases were resection specimen and all others were incisional biopsies. The largest tumor size was 5.5 cm in one of the resections. The incisional/fragmented tissues were all less than 5 cm in aggregate. Most cases showed predominance of nodular growth and a minor component of spindle cell population along with lymphangiomatosis like vascular channels, with evidence of microthrombi in 2 cases. Few multinucleated giant cells were observed in 2 cases. None of the cases exhibited significant nuclear atypia or mitotic activity. One of the cases arising in dermis showed underlying bone involvement. HHV8 was negative in 7/7 cases. Conclusions KHE can also involve adult population and it should always be considered in the differential diagnoses of a vascular lesion. Presence of multinucleated giant cells is a rare finding. Knowledge about histological features and potential mimics is helpful in avoiding misdiagnosis.

Computer-aided diagnostics is helping the medical experts for fast diagnostics, using machine learning and representation learning techniques. Various types of diagnostics are using the assistance of machine learning approaches including endoscopy. In this paper, a transfer learning based bagging approach is investigated for endoscopy images analysis. Bagging is used to fine-tune several instances of the deep learning model with 70% of data in each bag. These all models of deep learning are combined to generate a single prediction using majority voting and neural-network-based decision approach. The best approach resulted in an F1-score of 0.60 on the EndoTech 2020 dataset having 23 abnormalities in the GI-Tract.

Historically, the database search algorithms have been the de facto standard for inferring peptides from mass spectrometry (MS) data. Database search algorithms deduce peptides by transforming theoretical peptides into theoretical spectra and matching them to the experimental spectra. Heuristic similarity-scoring functions are used to match an experimental spectrum to a theoretical spectrum. However, the heuristic nature of the scoring functions and the simple transformation of the peptides into theoretical spectra, along with noisy mass spectra for the less abundant peptides, can introduce a cascade of inaccuracies. In this paper, we design and implement a Deep Cross-Modal Similarity Network called SpeCollate, which overcomes these inaccuracies by learning the similarity function between experimental spectra and peptides directly from the labeled MS data. SpeCollate transforms spectra and peptides into a shared Euclidean subspace by learning fixed size embeddings for both. Our proposed deep-learning network trains on sextuplets of positive and negative examples coupled with our custom-designed SNAP-loss function. Online hardest negative mining is used to select the appropriate negative examples for optimal training performance. We use 4.8 million sextuplets obtained from the NIST and MassIVE peptide libraries to train the network and demonstrate that for closed search, SpeCollate is able to perform better than Crux and MSFragger in terms of the number of peptide-spectrum matches (PSMs) and unique peptides identified under 1% FDR for real-world data. SpeCollate also identifies a large number of peptides not reported by either Crux or MSFragger. To the best of our knowledge, our proposed SpeCollate is the first deep-learning network that can determine the cross-modal similarity between peptides and mass-spectra for MS-based proteomics. We believe SpeCollate is significant progress towards developing machine-learning solutions for MS-based omics data analysis. SpeCollate is available at https://deepspecs.github.io/.

The present spreading out of big data found the realization of AI and machine learning. With the rise of big data and machine learning, the idea of improving accuracy and enhancing the efficacy of AI applications is also gaining prominence. Machine learning solutions provide improved guard safety in hazardous traffic circumstances in the context of traffic applications. The existing architectures have various challenges, where data privacy is the foremost challenge for vulnerable road users (VRUs). The key reason for failure in traffic control for pedestrians is flawed in the privacy handling of the users. The user data are at risk and are prone to several privacy and security gaps. If an invader succeeds to infiltrate the setup, exposed data can be malevolently influenced, contrived, and misrepresented for illegitimate drives. In this study, an architecture is proposed based on machine learning to analyze and process big data efficiently in a secure environment. The proposed model considers the privacy of users during big data processing. The proposed architecture is a layered framework with a parallel and distributed module using machine learning on big data to achieve secure big data analytics. The proposed architecture designs a distinct unit for privacy management using a machine learning classifier. A stream processing unit is also integrated with the architecture to process the information. The proposed system is apprehended using real-time datasets from various sources and experimentally tested with reliable datasets that disclose the effectiveness of the proposed architecture. The data ingestion results are also highlighted along with training and validation results.

Malignant eccrine spiradenoma (MES) is an exceedingly rare skin adnexal tumor that arises from pre-existing benign eccrine spiradenoma (BES). MES tumors show a wide spectrum of morphological features, posing a diagnostic challenge to the pathologist. Sarcomatous (heterologous) elements are seen in a few of these tumors, further complicating the morphological picture. We herein describe a case of a 66-year-old male who presented with a recently enlarging, ulcerated, nodular skin lesion over the right leg that had been present for the last 25 years. The patient underwent wide local excision of the tumor. Microscopic examination revealed a neoplastic lesion comprising benign and malignant components. The carcinomatous component showed features of infiltrating adenocarcinoma, not otherwise specified, whereas the sarcomatous component showed predominant osteosarcomatous and focal chondrosarcomatous differentiation. The benign component showed morphological and immunohistochemical features of BES. No adjuvant treatment was administered. The patient was alive and disease-free for 14 months, after which he was lost to follow-up. Careful identification and knowledge related to histological diversity are keys to the correct diagnosis of this rare tumor. MESs are potentially aggressive tumors, and therefore, close long-term follow-up should be maintained.

Objectives: Hospital readmission has become a focus of national attention as a potential indicator of healthcare quality and has a significant financial impact on healthcare system. Limited data is available regarding readmissions to Internal Medicine specialty from our sub-continent. It is, therefore, essential to determine the frequency and factors leading to readmissions, in order to avoid preventable readmissions and improve quality of healthcare provision. Methods: This retrospective study reviewed adult discharges from Internal Medicine specialty between October 2018 and February 2019 at Aga Khan University Hospital. Out of 1,835 discharges, 491 were randomly selected after excluding expiries. The frequency, factors and outcomes of readmission were noted. The studied outcomes included length of stay and in-hospital mortality. Results: Out of 491 patients, 15.3% were readmitted within 30-days of their discharge. Most of the readmitted patients were females (56%) and elderly with a mean age 67.1±14.9 years. Majority of the patient who got readmitted had multi-morbidities (68%) and were of functional Class-II (39%).The mean length of stay for index and readmission was between 4-7days. Eighty-percent readmissions were discharged as planned, 13% on request and seven-percent left against medical advice in their index admission. The most common causes of readmission were persistence of symptoms (43%) and nosocomial infection (29%). Avoidable causes included hospital-associated pneumonia, urinary tract infections and septic shock. Mortality in readmitted patients was 12%. Conclusions: The causes of readmission is multi-factorial, including advanced age, multi-morbidities, persistence of symptoms and nosocomial infections. Early follow-ups should be advised to prevent avoidable readmissions. doi: https://doi.org/10.12669/pjms.37.7.3575 How to cite this:Fatima S, Shamim S, Raffat S, Tariq M. Hospital readmissions in Internal Medicine Specialty: Frequency, associated factors and outcomes. Pak J Med Sci. 2021;37(7):---------. doi: https://doi.org/10.12669/pjms.37.7.3575 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Encapsulated Papillary Carcinoma (EPC) is a rare breast tumor with excellent prognosis. Treatment and stage of EPC is influenced by invasion and high nuclear grade. Our aim was to study the clinicopathological features of EPC, especially high grade tumors and to compare the features of invasive and non-invasive tumors. We reviewed clinicopathological features of 25 cases of EPC diagnosed at our institution from 2006 till 2020. Patients' age ranged from 21 to 75 years (median 55 years). Tumor size ranged from 1 to 9 cm (median 3.5 cm). Overall, invasion was present in 44% cases. High nuclear grade was observed in 24% cases. Majority of these high grade tumors were below 40 years. All of these tumors were 4 cm or larger in size. Two third of these tumors were invasive. Hormone receptor negativity and lymph node involvement was observed in 1 out of 3 cases, when performed. Clinicopathological and histological features of invasive and non-invasive tumors were compared and only lymph node involvement was found to be significantly more frequent in invasive tumors (p = 0.049). Median follow up duration was 18 months. All patients were alive and disease free except for a single patient who died of cerebrovascular accident. EPC has excellent clinical course. Invasion and high nuclear grade should be carefully searched for as these features determine tumor stage and treatment.

Background Papillary lesions of the breast constitute a heterogeneous group ranging from non-malignant papillomas to papillary carcinoma. While surgical excision is recommended for atypical papilloma or papillary DCIS/ carcinoma on core biopsy, controversy persists in the management of benign papillomas which are diagnosed with core needle biopsy (CNB) since there are variable reported rates for tumor upgrade. The purpose of this study was to determine the outcome of papillary lesions of the breast diagnosed at image-guided CNB, after surgical excision or follow-up, and to identify potential predictors of high-risk lesions/malignancy on imaging. Materials and methods We retrospectively identified 52 non-malignant papillary lesions on core biopsy between January 2012 and June 2018. The outcome of surgical excision, as well as clinical and imaging features of these lesions, were assessed. The final histologic upgrade was recorded, and variables were compared between benign and atypical lesions on core biopsy as well as between upgraded and non-upgraded lesions after surgical excision. Results Thirty-six lesions out of 52 lesions were benign papillomas on core biopsy, while 16 were papillary lesions with ADH/DCIS. All of these lesions except four benign papillomas were excised. Of the 32 benign papillomas excised, 7 were upgraded to papilloma with ADH/DCIS and one to DCIS with the focus of invasion. Among the 16 atypical lesions excised, one was upgraded to papillary DCIS with a final upgrade rate of 17.3%. There was no statistically significant clinical or imaging feature among those that were upgraded on excision from those that were not upgraded. Conclusion Non-malignant papillary lesions have a significant upgrade rate. There are no reliable clinical or imaging features that can pre-surgically predict upgrade. Therefore, surgical excision of all papillary lesions is recommended for definitive diagnosis.

Gonadoblastoma is a rare ovarian neoplasm which belongs to “germ cell-sex cord-stromal tumor” category. This tumor is frequently associated with invasive germ cell malignancy. It commonly arises in dysgenetic gonads of young individuals who are phenotypically females but possess 46XY karyotype. It has been rarely reported in females with normal phenotype and genotype. We herein describe a case of 10-year-old female who presented with abdominal pain, abdominal distention and fever. CT scan of the abdomen and pelvis revealed bilateral ovarian masses, ascites and pelvic and para-aortic lymphadenopathy. Serum lactate dehydrogenase levels were also elevated. She underwent left salpingo-oophorectomy, right ovarian biopsy, omentectomy and para-aortic lymphadenopathy. Microscopically, tumor showed in situ and invasive components. In situ component was arranged in nests and lobules formed by immature sertoli cells forming acini and encircling large polygonal primitive germ cells. Immature sertoli cells were positive for immunohitochemical (IHC) stains cytokeratin AE1/AE3, inhibin and calretinin, while germ cells were positive for SALL4, Oct 3/4, placental alkaline phosphatase (PLAP) and CD117. Invasive component was arranged in sheets of large-sized, polygonal-shaped primitive germ cells which were also positive for SALL4, Oct 3/4, PLAP and CD117 IHC stains. Hence, the diagnosis of “gonadoblastoma with dysgerminoma” was made. The tumor was limited to both ovaries. Cytogenetic analysis of peripheral blood revealed normal female 46XX karyotype. The patient received two cycles of adjuvant chemotherapy and was then lost to follow-up. We conclude that gonadoblastoma, although rare, should be considered as a differential diagnosis in ovarian tumors of young females. Invasive germ cell component should always be carefully searched for as it guides about treatment and predicts prognosis.

Nipple adenoma (NA) is a rare benign breast neoplasm that seldom co-exists with breast carcinoma (BC). Majority of these BC are separate from NA, and their origin from NA is an extremely rare event. We herein describe a case of 65-year-old female who had a painless lump for 15 years which increased in size and ulcerated for last six months. Microscopic examination of the wedge biopsy of nipple showed features of NA at superficial aspect and invasive carcinoma from it at the deeper aspect. The patient underwent mastectomy and axillary clearance, which revealed a 4-cm invasive breast carcinoma, no special type with axillary lymph node involvement. The patient received adjuvant chemotherapy, radiotherapy and adjuvant hormonal therapy. The patient is alive and disease-free after 36 months. NA should be carefully evaluated for co-existent BC because it completely changes the treatment plan and prognosis.

OBJECTIVE To determine the reliability, validity, feasibility, acceptability and perceived educational impact of WATCH: Warwick Assessment insTrument for Clinical teacHing among doctors in Pakistan. STUDY DESIGN Cross-sectional research study. PLACE AND DURATION OF STUDY The College of Physicians and Surgeons, Pakistan, from September 2018 to August 2019. METHODOLOGY Postgraduate trainees were asked to rate the clinical teaching sessions, using WATCH, which consists of 15 items. Percentage was used to calculate gender and participation from different specialties. Inter-item correlations of 15 items with individual mean scores, standard deviations and Cronbach's Alpha were reported, including Friedman test, in order to observe the scores across multiple conditions. The Hotelling's T2 test was used to test whether the answers provided by the study participants to the questionnaire were equal. Construct validity was determined using factor analysis while feasibility, acceptability, and educational impact was evaluated by seeking participants' feedback on five semi-structured questions. RESULTS More than 80% ranked WATCH from good to excellent. Oveall 8 items were perceived as excellent, while 7 items received rating of good. Inter-item correlation ranged from 0.61 to 0.81. Cronbach Alpha was reported to be 0.975, with significant difference in mean scores of different items (Friedman's Chi-Square=4285.54; p

Pseudoangiomatous stromal hyperplasia (PASH) is a benign mesenchymal proliferative lesion of the breast, often an incidental finding on breast biopsy specimens and rarely presents as a palpable lump. The case being reported is interesting as a lactating female presented with gross left breast enlargement due to a huge firm mass with skin thickening and palpable left axillary lymph nodes. A provisional diagnosis of left breast malignancy was made and the patient extensively worked up with ultrasound, CT scan, bone scan and core biopsy. The histopathology, however, revealed PASH of the breast. There was no invasive or in situ malignancy. The patient was successfully managed conservatively.

Papillary urothelial carcinoma (PUC) is the most common malignant tumor of the urinary bladder. Urothelial tumors are notorious for frequent recurrences and follow a chronic relapsing course in most of the patients. In Pakistan, the incidence of PUC is showing a rising trend. Various immunohistochemical (IHC) markers including androgens have been studied as prognostic and predictive markers in PUC with conflicting results. Androgen is a steroid-based sex hormone and plays an important role in different body organs such as urinary bladder, prostate, muscles, and brain. We aimed to investigate the role of the IHC expression of androgen receptor (AR) as a predictor of recurrence in papillary urothelial carcinoma patients. Eighty-four patients were included in the study. Tissues from the biopsy specimens of these cases were stained with antibodies against AR; 17% of the cases demonstrated a positive AR IHC expression. The expression was slightly more common in low-grade carcinoma. The AR expression was not significantly associated with clinicopathological features. Recurrence was observed in 49% of the cases, and it was significantly more common in AR-negative cases (p-value: 0.025). Eighteen out of 19 patients who died of disease were AR- negative, but no statistical significance was observed. We conclude that the IHC expression of AR can be used as a predictive marker for PUC as it correlates with the recurrence rate.

This paper investigates how capital requirements and bank competition affect banks' financial soundness in the Middle East and North Africa (MENA) region. We test the hypothesis that regulatory capital positively impacts the risk-taking behavior of Islamic and conventional banks in the MENA region. The analysis indicates that the capital adequacy ratio has no significant influence on the credit risk of Islamic banks; however, market competition does play a significant role in shaping the risk behavior of these institutions. We report the opposite result for conventional banks – an increase in the minimum capital requirements is associated with an increase in their risk level. We also find that Islamic and conventional banks experience a non-linear relationship between market concentration and bank risk. Our findings inform regulatory authorities concerned with improving banking sector’s stability in the MENA region to strengthen their policies to force banks to better align with capital requirements which is highly important during the COVID-19 pandemic.

The proliferation of sensor-based applications in healthcare has given rise to Internet of Health Things (IoHT) that improves patient safety, staff morale, and operational efficiency. Edge-fog computing has seen significant development in recent years and supports the association of various intelligent things with sensors for establishing smooth data transfer. However, it becomes challenging for edge-fog computing to tackle diverse IoHT settings such as efficient disease management, emergency response management, etc. The key limitation of existing architectures is the restricted scalability and inability to meet the demands of hierarchical computing environments for IoHT. This is because latency-sensitive applications often require large quantities of data to be measured and transferred to the data centers, which causes delay and reduced output. This research proposes a novel edge-fog computing framework for the convergence of machine learning ensemble with edge-fog computing. The proposed architecture delivers healthcare as a fog system that handles data from different sources to manage the diseases effectively. The proposed framework is used for the real-life implementation and automatic detection of gliomas diseases. Glioma is a kind of tumor, which ensues in the spinal cord and a portion of the brain. Glioma instigates in the glial cells that surround the nerve cells. The proposed edge-fog framework efficiently manages the real-time data related to gliomas. This framework is configured for specific operating modes including diverse edge-fog scenarios, different user requirements, quality of service, precision, and predictive accuracy. The proposed framework is evaluated using real-time datasets from various sources and experimentally tested with reliable datasets that disclose the effectiveness of the proposed architecture. The performance of the proposed model is evaluated in terms of power consumption, latency, accuracy, and execution time, respectively.

Objective The status of axillary lymph nodes is one of the most important prognostic factors in patients with breast cancer. A precise noninvasive evaluation of axillary lymph node status preoperatively, although challenging, is vital for optimization of the treatment plan for patients. The objective of our study was to assess the utility of ultrasound and mammography in detecting the absence of axillary lymph nodal metastasis in patients of breast cancer, taking histopathology as gold standard. Methods A cross-sectional study was conducted in the Department of Radiology, Aga Khan University Hospital, Karachi. All female patients between 20 and 95 years of age with a known diagnosis of breast cancer with mammographic and ultrasound imaging done at our institute were included. Patients with abnormal lymph nodes on mammography or on ultrasound, patients already operated for breast cancer, patients who already underwent axillary lymph node dissection and those whose histopathology reports were not available or who did not undergo surgery were excluded. Results A total of 262 women with breast carcinoma who had both ultrasound and mammography done and also had surgery performed at our institution were included. At final surgical pathology, a total of 45 of the 262 patients (17.2%) with breast carcinoma had one or more positive lymph nodes. Out of the total 262 patients, 217 patients were found to be true negatives as they had absent axillary nodal metastasis on imaging as well as on histopathology. In all, 45 out of 262 patients were found to be false negatives as they had absent axillary nodal metastasis on imaging; however, they were found to be positive for metastasis on histopathology. The negative predictive value was 82.8%. Patient age was considered as a factor that may influence the outcome of results; the patients were stratified into age ranges seven groups with the age range of 10 years, ranging from 26 to 95 years. Chi-square test showed a p-value of 0.148, which showed no significant difference in the effect of age on diagnosing the absence of metastasis by ultrasound and mammography. Conclusion Our study shows that ultrasound and mammography even when used in combination cannot safely exclude axillary metastasis and thus cannot eliminate the need for sentinel node biopsy.

Due to increased usage of digital technologies in all sectors and in almost all day to day activities to store and pass information, Handwriting character recognition has become a popular subject of research. Handwriting remains relevant, but people still want to have Handwriting copies converted into electronic copies that can be communicated and stored electronically. Handwriting character recognition refers to the computer's ability to detect and interpret intelligible Handwriting input from Handwriting sources such as touch screens, photographs, paper documents, and other sources. Handwriting characters remain complex since different individuals have different handwriting styles. This paper aims to report the development of a Handwriting character recognition system that will be used to read students and lectures Handwriting notes. The development is based on an artificial neural network, which is a field of study in artificial intelligence. Different techniques and methods are used to develop a Handwriting character recognition system. However, few of them focus on neural networks. The use of neural networks for recognizing Handwriting characters is more efficient and robust compared with other computing techniques. The paper also outlines the methodology, design, and architecture of the Handwriting character recognition system and testing and results of the system development. The aim is to demonstrate the effectiveness of neural networks for Handwriting character recognition.

The cardiovascular system (CVS) includes both the heart, i.e. endocardium, myocardium and pericardium, and vessels. Biopsies from these organs are relatively uncommon, and most cases actually originate from autopsies. Disease spectrum ranges from ‘inflammatory’, ‘ischaemic’ and ‘degenerative’ with a handful of primary and metastatic neoplasms involving the heart. Various distinct types of vasculopathies are also sometimes seen in clinical practice. In this chapter, morphological features on light microscopy and utility of special and immunohistochemical (IHC) stains, which are helpful in the workup of these lesions, are described in a compact fashion.

Organizations invest intense resources in their product development processes. This paper aims to create a knowledge environment using trade-off curves during the early stages of the set-based concurrent engineering (SBCE) process of an aircraft jet engine for a reduced noise level at takeoff. Data is collected from a range of products in the same family as the jet engine. Knowledge-based trade-off curves are used as a methodology to create and visualize knowledge from the collected data. Findings showed that this method provides designers with enough confidence to identify a set of design solutions during the SBCE applications.

Breast biopsies are one of the most frequently submitted biopsies in any diagnostic surgical pathology practice as breast cancer is by and large the most common cancer among women worldwide. This chapter covers morphologic light microscopic features of most if not all disease entities. Focus is on conventional H&E stained slides, however where necessary immunohistochemical (IHC) stains are also included. Besides unequivocal benign and malignant neoplasms, a major challenge for pathologists is the interpretation of ‘borderline lesions’, where most controversies and difference of opinion arise in this era where most biopsies are small needle core biopsies with limited yield. This requires robust diagnostic criteria laid down by experts to minimize subjectivity. It is hoped that this resource will be helpful for trainees as well as for general diagnostic pathologists as a quick reference in their day-to-day practice and diagnostic challenges.

Bone and soft tissue pathology is considered to be one of the most challenging areas of diagnostic pathology. It requires not only radiologic films at hand but also frequent use of ancillary tests like special and immunohistochemical (IHC) stains. Many chromosomal translocations and other chromosomal abnormalities are hallmark of several of these tumours which require molecular genetic testing usually by using techniques as FISH. Hence an attempt is made to encompass not only light microscopic features and pattern analysis but also judicious use of special and IHC stains as well as FISH to describe and diagnose a distinct entity compatible with clinical presentation and radiologic findings.