Revolutionizing tumor detection and classification in multimodality imaging based on deep learning approaches: methods, applications and limitations.

BACKGROUND: The emergence of deep learning (DL) techniques has revolutionized tumor detection and classification in medical imaging, with multimodal medical imaging (MMI) gaining recognition for its precision in diagnosis, treatment, and progression tracking. OBJECTIVE: This review comprehensively examines DL methods in transforming tumor detection and classification across MMI modalities, aiming to provide insights into advancements, limitations, and key challenges for further progress. METHODS: Systematic literature analysis identifies DL studies for tumor detection and classification, outlining methodologies including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and their variants. Integration of multimodality imaging enhances accuracy and robustness. RESULTS: Recent advancements in DL-based MMI evaluation methods are surveyed, focusing on tumor detection and classification tasks. Various DL approaches, including CNNs, YOLO, Siamese Networks, Fusion-Based Models, Attention-Based Models, and Generative Adversarial Networks, are discussed with emphasis on PET-MRI, PET-CT, and SPECT-CT. FUTURE DIRECTIONS: The review outlines emerging trends and future directions in DL-based tumor analysis, aiming to guide researchers and clinicians toward more effective diagnosis and prognosis. Continued innovation and collaboration are stressed in this rapidly evolving domain. CONCLUSION: Conclusions drawn from literature analysis underscore the efficacy of DL approaches in tumor detection and classification, highlighting their potential to address challenges in MMI analysis and their implications for clinical practice.

Quantitative Ethnobotanical Study of Indigenous Knowledge on Medicinal Plants Used by the Tribal Communities of Gokand Valley, District Buner, Khyber Pakhtunkhwa, Pakistan.

The current study on the traditional use of medicinal plants was carried out from February 2018 to March 2020, in Gokand Valley, District Buner, Pakistan. The goal was to collect, interpret, and evaluate data on the application of medicinal plants. Along with comprehensive notes on individual plants species, we calculated Use Value (UV), Relative Frequency of Citation (RFC), Use Report (UR), Fidelity Level (FL), Informant Consensus Factor (FCI), as well as Family Importance Value (FIV). During the current study, a total of 109 species belonging to 64 families were reported to be used in the treatment of various ailments. It included three families (four species) of Pteridophytes, 58 families (99 species) of angiosperm, one family (three species) of Gymnosperms, and two families (three species) of fungi. The article highlights the significance of domestic consumption of plant resources to treat human ailments. The UV varied from 0.2 (Acorus calamus L.) to 0.89 (Acacia modesta Wall.). The RFC ranged from 0.059 (Acorus calamus L. and Convolvulus arvensis L.) to 0.285 (Acacia modesta Wall.). The species with 100% FL were Acacia modesta Wall. and the fungus Morchella esculenta Fr., while the FCI was documented from 0 to 0.45 for gastro-intestinal disorders. The conservation ranks of the medicinal plant species revealed that 28 plant species were vulnerable, followed by rare (25 spp.), infrequent (17 spp.), dominant (16 spp.), and 10 species endangered. The traditional use of plants needs conservation strategies and further investigation for better utilization of natural resources.

Computer-aided osteoporosis detection from DXA imaging.

BACKGROUND AND OBJECTIVE: Osteoporosis is a skeletal disease caused by a high rate of bone tissue loss, and it is a major cause of bone fracture. In contemporary society, osteoporosis is more common than cancer and stroke and results in a higher rate of morbidity and mortality in the human population. Osteoporosis can conclusively be diagnosed with dual energy X-ray absorptiometry (DXA). In this study, we propose a computer-aided osteoporosis detection (CAOD) technique that automatically measures bone mineral density (BMD) and generates an osteoporosis report from a DXA scan. METHODS: The CAOD model denoise and segments DXA images using a non-local mean filter, Machine learning pixel label random forest respectively, and locates regions of interest with higher accuracy. Pixel label random forest classifies a pixel either bone or soft tissue; then contours are extracted from binary image to locate regions of interest and calculate BMD from bone and soft tissues pixels. Mean standard deviation and correlation coefficients statistical analysis were used to evaluate the consistency and accuracy of BMD measurements. RESULTS: During a consistency test of BMD measurements using three consecutive scans from Computerized Imaging Reference Systems' Bona Fide Phantom (CIRS-BFP) for the spine, the CAOD model showed an averaged standard deviation of 0.0029 while the standard deviation from manual measurements on the same data set by three different individuals was recorded as 0.1199. During another correlation study of BMD measurements evaluating real human scan images by the CAOD model versus manual measurement, the model scored a correlation coefficient of R2 = 0.9901 while the CIRS-BFP study scored a correlation coefficient of R2 = 0.9709. CONCLUSIONS: The CAOD model increases the preciseness and accuracy of BMD measurements. This CAOD method will help clinicians, untrained DXA operators, and researchers (medical scientists, doctors, and bone researchers) use the DXA system with reliable accuracy and overcome workload challenges. It will also improve osteoporosis diagnosis from DXA systems and increase system performance and value.

Automatic hip geometric feature extraction in DXA imaging using regional random forest.

BACKGROUND: Hip fracture is considered one of the salient disability factors across the global population. People with hip fractures are prone to become permanently disabled or die from complications. Although currently the premier determiner, bone mineral density has some notable limitations in terms of hip fracture risk assessment. OBJECTIVES: To learn more about bone strength, hip geometric features (HGFs) can be collected. However, organizing a hip fracture risk study for a large population using a manual HGFs collection technique would be too arduous to be practical. Thus, an automatic HGFs extraction technique is needed. METHOD: This paper presents an automated HGFs extraction technique using regional random forest. Regional random forest localizes landmark points from femur DXA images using local constraints of hip anatomy. The local region constraints make random forest robust to noise and increase its performance because it processes the least number of points and patches. RESULTS: The proposed system achieved an overall accuracy of 96.22% and 95.87% on phantom data and real human scanned data respectively. CONCLUSION: The proposed technique's ability to measure HGFs could be useful in research on the cause and facts of hip fracture and could help in the development of new guidelines for hip fracture risk assessment in the future. The technique will reduce workload and improve the use of X-ray devices.

Femur segmentation in DXA imaging using a machine learning decision tree.

BACKGROUND: Accurate measurement of bone mineral density (BMD) in dual-energy X-ray absorptiometry (DXA) is essential for proper diagnosis of osteoporosis. Calculation of BMD requires precise bone segmentation and subtraction of soft tissue absorption. Femur segmentation remains a challenge as many existing methods fail to correctly distinguish femur from soft tissue. Reasons for this failure include low contrast and noise in DXA images, bone shape variability, and inconsistent X-ray beam penetration and attenuation, which cause shadowing effects and person-to-person variation. OBJECTIVE: To present a new method namely, a Pixel Label Decision Tree (PLDT), and test whether it can achieve higher accurate performance in femur segmentation in DXA imaging. METHODS: PLDT involves mainly feature extraction and selection. Unlike photographic images, X-ray images include features on the surface and inside an object. In order to reveal hidden patterns in DXA images, PLDT generates seven new feature maps from existing high energy (HE) and low energy (LE) X-ray features and determines the best feature set for the model. The performance of PLDT in femur segmentation is compared with that of three widely used medical image segmentation algorithms, the Global Threshold (GT), Region Growing Threshold (RGT), and artificial neural networks (ANN). RESULTS: PLDT achieved a higher accuracy of femur segmentation in DXA imaging (91.4%) than either GT (68.4%), RGT (76%) or ANN (84.4%). CONCLUSIONS: The study demonstrated that PLDT outperformed other conventional segmentation techniques in segmenting DXA images. Improved segmentation should help accurate computation of BMD which later improves clinical diagnosis of osteoporosis.

Denoising images of dual energy X-ray absorptiometry using non-local means filters.

BACKGROUND: In general, the image quality of high and low energy images of dual energy X-ray absorptiometry (DXA) suffers from noise due to the use of a small amount of X-rays. Denoising of DXA images could be a key process to improve a bone mineral density map, which is derived from a pair of high and low energy images. This could further improve the accuracy of diagnosis of bone fractures and osteoporosis. OBJECTIVE: This study aims to develop and test a new technology to improve the quality, remove the noise, and preserve the edges and fine details of real DXA images. METHODS: In this study, a denoising technique for high and low energy DXA images using a non-local mean filter (NLM) was presented. The source and detector noises of a DXA system were modeled for both high and low DXA images. Then, the optimized parameters of the NLM filter were derived utilizing the experimental data from CIRS-BFP phantoms. After that, the optimized NLM was tested and verified using the DXA images of the phantoms and real human spine and femur. RESULTS: Quantitative evaluation of the results showed average 24.22% and 34.43% improvement of the signal-to-noise ratio for real high and low spine images, respectively, while the improvements were about 15.26% and 13.55% for the high and low images of the femur. The qualitative visual observations of both phantom and real structures also showed significantly improved quality and reduced noise while preserving the edges in both high and low energy images. Our results demonstrate that the proposed NLM outperforms the conventional method using an anisotropic diffusion filter (ADF) and median techniques for all phantom and real human DXA images. CONCLUSIONS: Our work suggests that denoising via NLM could be a key preprocessing method for clinical DXA imaging.

PERFORATED DUODENAL ULCER ASSOCIATED WITH SITUS INVERSUS AND DEXTROCARDIA.

A 32 years old gentleman, presented in emergency department, with complaints of sudden onset of severe upper abdominal pain, associated with nausea and vomiting. He was a known case of acid peptic disease. His abdominal examination showed signs of peritonitis. X-ray chest showed pneumoperitoneum, with dextrocardia. Ultrasound showed situs inversus. Exploration confirmed the diagnosis of perforated ulcer and situs inversus. Grahm's patch repair of perforation was done. His postoperative recovery was smooth.

Dual Ectopy of Thyroid Tissue: Intratracheal and in Nasopharynx, Associated with Normally Located Thyroid.

Dual ectopic thyroid tissue associated with normally located thyroid, is a rare anomaly. We report a case of 42 years lady, who underwent left hemithyroidectomy for nodules in the left lobe of thyroid 15 years back. She presented with right side neck swelling and dyspnea for 6 months. Right hemithyroidectomy was done for multinodular goiter involving right lobe of thyroid. She developed stridor on the first postoperative day. CT scan and MRI showed intratracheal ectopic thyroid tissue. Ectopic thyroid tissue was removed by dividing cricoid cartilage, and trachea was repaired. Radio iodine scan after 6 weeks showed another ectopic thyroid tissue in nasopharynx. Patient refused for radio iodine ablation and, she was given thyroxin replacement.

LAPAROSCOPIC CHOLECYSTECTOMY: OUTCOME OF FIRST 202 CASES IN A DISTRICT HOSPITAL IN GILGIT.

BACKGROUND: The incidence of gall stone disease is on the rise in Gilgit Baltistan. The objectives of the study were to assess the outcome of laparoscopic cholecystectomy in terms of conversion rate and postoperative morbidity, in The Aga Khan Medical Centre Gilgit. METHODS: It was descriptive case series. All patients that underwent laparoscopic cholecystectomy between June 2009 to May 2014 were included. The data was collected prospectively. Demographic features, operative time, and hospital stay were studied. Postoperative complications were documented and evaluated according to outcome measures (bile duct injuries, morbidity, mortality, conversion rates, wound infections). RESULTS: A total of 202 consecutive patients were enrolled with a mean age of 49±15 years. There were 164 (81%) female and 38(19%.) male patients. Twenty nine (15%) patients had hypertension, 51 (25%) patients had diabetes mellitus as comorbid conditions. The mean operative time was 54±21 minutes. The operative time was longer in 52 (26%) patients. Three patients (1.5%) required conversion to open cholecystectomy due to obscured anatomy in the area of Calot's triangle, and empyema gallbladder. The mean hospital stay was 2±0.7 days. No common bile duct injury, solid organ or bowel injury occurred in this study. The mean follow up duration was 30±15 months. Postoperative complications include, port site infection in 8 (2%) patient, chest infection in 5 (2.4%) patients, and one (0.5%) patient had myocardial infarction. There was no mortality reported in this group of patients. CONCLUSION: Laparoscopic cholecystectomy is a safe procedure with advantages of decreased wound infection, less pain, decreased hospital stay, and early recovery.

Laparoscopic repair of ventral hernia.

OBJECTIVE: To determine the outcome of patients in terms of wound complications and follow-up, after laparoscopic ventral hernia repair. STUDY DESIGN: A case series study. PLACE AND DURATION OF STUDY: Department of Surgery, Dubai Hospital, Dubai, from January 2007 to December 2011. METHODOLOGY: All patients above 13 years of age, who underwent laparoscopic ventral hernia repair, were included. A proceed dual mesh was used for repair of hernia. Multiple parameters were analyzed, including demographic features, presentations, co-morbid conditions, duration of hernia, and defect size. The duration of surgery, postoperative complications, and follow-up of these patients in terms of suture site pain and recurrence of hernia were also analyzed. The data was expressed as frequency, percentages and mean ± standard deviation of values. RESULTS: There were 27 patients with mean age of 47 ± 10.3 years, including 20 female (74.07%) and 7 male (25.95%)patients. All patients presented with abdominal wall swelling. The hernia was partially reducible in 12 patients (44.44%), and completely reducible in 15 patients (55.55%). Seven patients (25.95%) had hypertension, 4 (14.81%) had ischaemic heart disease, and 4 (14.81%) had obesity as co-morbid conditions. All patients underwent laparoscopic hernia repair with proceed dual mesh. The mean defect size of the hernia was 6 cm, and mean duration of surgery was 94 minutes. Early postoperative complications included, seroma in 3 patients (11.11%), and haematoma in one patient (3.70%). The mean follow-up was 23 months. Four patients (14.81) had pain at suture site. CONCLUSION: Laparoscopic repair is an appropriate approach for ventral hernia repair. It results in good repair and low wound complications in terms of haematoma and wound infection. There was no recurrence of hernia in this study.

Acute mesenteric ischemia: experience in a tertiary care hospital.

BACKGROUND: Acute mesenteric ischemia is an abdominal catastrophe. This has been described as a complex of diseases rather than a single clinical entity. The incidence in United States is 1 in 1000 hospital admissions. The objective of this descriptive study was to determine the clinical presentations and out come after surgery of patients with acute mesenteric ischemia. It was conducted at Dubai Hospital, Dubai, United Arab Emirates. METHODS: All patients having per operative or histopathological diagnosis of acute mesenteric ischemia from 2002 to 2006 were included. RESULTS: There were 16 patients in all. Their mean age was 51 years, 12 were male and 4 were female. Abdominal pain was present in 16 patients, vomiting in 12 and anorexia in 9 patients. Abdominal tenderness was present in 16 patients, abdominal distension and rebound tenderness in 12 patients. Five patients had hypertension, 4 had myocardial infarction and 4 had diabetes mellitus as risk factors. X-Ray abdomen was done in 13 patients, Ultrasound in 9 and CT Scan in one patient. Resection of bowel was done in 14 patients. Post operatively 5 patients developed pneumonia, 3 had wound dehiscence, 3 had sepsis, and 3 had Lower GI bleeding. Five patients were expired after surgery in the hospital. Four patients were lost to follow up. CONCLUSION: We should have a high index of suspicion for mesenteric ischemia in patients with unexplained abdominal pain. Early diagnosis and prompt surgical intervention improves the outcome.

Magnetic attraction: dual complications in a single case.

A case report of a 19-years-old mentally retarded girl, presented with abdominal pain, nausea and vomiting of two days duration. Physical examination revealed abdominal tenderness with guarding and rigidity. Exploratory laparotomy revealed the presence of magnets in the jejunum with perforation of jejunum and a resultant gastro jejunal fistula. The procedure involved enterotomy for removal of magnets, resection of the segment of the bowel with fistula and perforation and end-to-end anastomosis. The association of ingested magnets, leading to dual complications, is a rare case ever reported in Dubai.

Gastric stromal tumors: clinical presentations, diagnosis and outcome.

OBJECTIVE: To determine the clinical presentations, of gastric stromal tumors with diagnostic methods, pathology and outcome after surgery. STUDY DESIGN: A case series. DURATION AND SETTING: From January 1988 to December 2002 at The Aga Khan University Hospital, Karachi. PATIENTS AND METHODS: All patients of age 14 years and above, diagnosed histopathologically to have gastric stromal tumors were included. The data of these patients was collected retrospectively from January 1988 to December 1998, and prospectively from January 1999 to December 2002. All the patients were studied as a single group. RESULTS: There were 11 patients. Their mean age was 54 years, with 8 males and 3 females. Five patients presented with upper gastrointestinal bleeding, and 4 with lower gastrointestinal bleeding. Eight patients had pain in epigastrium and 2 had vomiting. Upper gastrointestinal endoscopy was done in all patients, and ultrasound was done in 4 patients. CT scan was done in 7 patients. Pre-operative diagnosis could be made in 6 patients. Only one patient had liver metastasis. Wedge resection was performed in 5 proximal gastrectomy with gastroesophageal anastomosis in 3, and partial gastrectomy with gastrojejunostomy in another 3 patients. The mean tumor size was 8.0 centimeters. Two patients had benign, 2 had intermediate and 7 had malignant tumors. The mean duration of follow-up was 41 months. Follow-up was completed in 8 patients, out of whom 6 were alive, and 2 patients expired due to other causes at the time of completion of this study. CONCLUSION: Gastric stromal tumors are uncommon. Larger gastric stromal tumors are usually symptomatic with gastrointestinal bleeding as a common presentation. Immunohistochemical techniques are required for the diagnosis. Complete surgical resection is the curative therapy.

Management outcome of residual common bile duct stones at Aga Khan University Hospital.

BACKGROUND: The frequency of residual common bile duct (CBD) stones in patients with previous cholecystectomy ranges from 2-10%, and a minimally invasive approach is generally recommended for these patients. This study reviews the experience in the management of residual CBD stones at the Aga Khan University Hospital. METHODS: All adult patients diagnosed to have residual CBD stones, from 1993 to 2001, were identified and analyzed in terms of the treatment modality utilized and its associated complications. The role of endoscopic sphincterotomy was particularly analyzed in our set-up. RESULTS: The study population consisted of 66 patients. The mean age was 52 (range:18-84 years) years with a female to male ratio of 51:15. The primary mode of management consisted of endoscopic sphincterotomy in 61 patients (92.5%). The initial clearance rate for these patients was 75%, while the remaining 25% required ancillary procedures to achieve a complete clearance. Procedure related complications were observed in 17 (28%) patients, with zero mortality. The other 5 patients (7.5%) underwent an open choledochotomy as a primary procedure with no further complications. CONCLUSIONS: Endoscopic sphincterotomy is the most frequent treatment modality used for the management of residual CBD stones at our hospital. Although initial success rate seems low, the fact that endoscopic sphincterotomy is a less invasive procedure justifies its preferential utilization. The study does not compare the results of endoscopic management with open surgery, as the number of patients managed by open choledochotomy is very small.

Surgical palliation for unresectable pancreatic carcinoma.

OBJECTIVES: To review the results of surgical palliation for unresectable pancreatic carcinoma, and to analyze the morbidity and mortality associated with the surgical procedure. The reasons for readmission after discharge from the hospital were also analyzed. METHODS: A retrospective study from 1995 to 2001 was done on 30 patients with pancreatic cancer operated with palliative intent, or those explored with curative intent but histopathology revealed positive resection margins or lymph node metastasis. RESULTS: Twenty-five (83.25%) patients were above 50 years of age. There were 16 (53.28%) male, and 14 (46.62%) females, 8 (26.64%) had diabetes mellitus, 2 (6.66%) chronic pancreatitis and 4 (13.32%) had smoking as risk factors. Twenty-three (76.59%) patients presented with jaundice, 18 (59.94%) with weight loss, 17 (56.61%) with epigastric pain, 15 (49.95%) with anorexia and 14 (46.62%) with vomiting. Whipple's procedure was performed in 9 (29.97%) patients, triple bypass in 13 (43.29%), choledochojejunostomy and gastrojejunostomy in 3 (9.99%) and gastrojejunostomy alone in 5 (16.65%) patients. Seven (23.31%) patients had preoperative ultrasonography, while CT Scan was done in 24 (79.92%) and ERCP in 8 (26.64%) patients. Histopathology showed positive resection margins in 9 (29.97%) patients and lymph node metastasis in 5 (16.65%) patients. Seventeen (56.61%) patients received less than 2 units of pack cells transfusion. Most of the patients remained admitted in the hospital between 20 to 30 days. Post-operatively, delayed gastric emptying was detected in 6 (19.98%) patients, cholangitis in 2 (6.66%), wound infection in 3 (9.99%), anastomotic leak in 2 (6.66%) and line sepsis in 2 (6.66%) patients. Three (9.99%) patients expired in hospital post operatively. The reasons for re-admission after discharge included abdominal pain in 9 (29.97%) patients, anemia in 3 (9.99%), intestinal obstruction in 3 (9.99%) and urinary tract infection in 2 (6.66%) patients. Follow up record was available for 22 (73.26%) patients. Six (19.98%) patients survived for 5 to 6 months and 9 (29.97%) had a survival between 7 to 10 months. CONCLUSION: A single surgical procedure can palliate all three symptoms associated with unresectable pancreatic carcinoma, and can be carried out with reasonable safety in selected patients. The commonest indication for re-admission is severe abdominal pain associated with advanced malignancy, hence chemical splanchiectomy may also be considered at the time of surgical exploration.

Liver cyst with inflammatory hepatic duct polyp.

This is a case report of a 50-year-old woman, who presented with upper abdominal pain, vomiting and two episodes of jaundice in five months. Computerised tomographic scan revealed a cyst in the 4th segment of the liver. Peroperatively it was found to be a simple liver cyst arising from the left hepatic duct. The cyst was excised. Peroperative cholangiogram revealed obstruction in the common hepatic duct. A polyp in the common hepatic duct, causing obstruction, was excised. The presence of liver cyst with hepatic duct polyp is a rare association.