Computational pathology: A survey review and the way forward.

Computational Pathology (CPath) is an interdisciplinary science that augments developments of computational approaches to analyze and model medical histopathology images. The main objective for CPath is to develop infrastructure and workflows of digital diagnostics as an assistive CAD system for clinical pathology, facilitating transformational changes in the diagnosis and treatment of cancer that are mainly address by CPath tools. With evergrowing developments in deep learning and computer vision algorithms, and the ease of the data flow from digital pathology, currently CPath is witnessing a paradigm shift. Despite the sheer volume of engineering and scientific works being introduced for cancer image analysis, there is still a considerable gap of adopting and integrating these algorithms in clinical practice. This raises a significant question regarding the direction and trends that are undertaken in CPath. In this article we provide a comprehensive review of more than 800 papers to address the challenges faced in problem design all-the-way to the application and implementation viewpoints. We have catalogued each paper into a model-card by examining the key works and challenges faced to layout the current landscape in CPath. We hope this helps the community to locate relevant works and facilitate understanding of the field's future directions. In a nutshell, we oversee the CPath developments in cycle of stages which are required to be cohesively linked together to address the challenges associated with such multidisciplinary science. We overview this cycle from different perspectives of data-centric, model-centric, and application-centric problems. We finally sketch remaining challenges and provide directions for future technical developments and clinical integration of CPath. For updated information on this survey review paper and accessing to the original model cards repository, please refer to GitHub. Updated version of this draft can also be found from arXiv.

Precision Medicine in Adult Obstructive Sleep Apnea and Home Diagnostic Testing: Caution in Interpretation of Home Studies Without Clinician Input Is Necessary.

Purpose: To assess the validity of home sleep apnea test directed diagnosis and treatment of obstructive sleep apnea (OSA) in a real-life clinical setting and establish the extent to which clinical evaluation alters diagnosis and therapeutic intervention, in the context of the evolving realm of precision medicine. Methods: Retrospective consecutive cohort study of 505 patients referred to a single center between 15th September 2015 to 14th September 2016, multidisciplinary specialist sleep clinic presenting with a home sleep apnea test prior to referral. We evaluated the effect of sleep medicine practitioner (SMP) and ear, nose, and throat surgeon (ENTS) review on patient diagnoses, disease severity, and management options in OSA. Results: Hundred and fifteen patients were included. Repeat evaluation with in-lab polysomnogram (PSG) was required in 46/115 (40.0%) of patients, of which 20/46 (43.5%) had OSA severity changed. Sleep medicine practitioner review decreased the need for repeat testing with formal in-lab PSG (p < 0.05) and increased patient acceptance of continuous positive airway pressure (CPAP) as a long-term management option for OSA. Sleep medicine practitioner/ENTS review resulted in discovery of a non-OSA related sleep disorder or change in OSA severity in 47.8% (55/115). Ear, nose, and throat surgeon review resulted in additional or changed diagnosis in 75.7% (87/115) of patients. Conclusion: In the clinical assessment and diagnosis of OSA, patients should be reviewed by medical practitioners with an interest in sleep disorders to better navigate the complexities of assessment, as well as the identification of co-morbid conditions.

Olfactory-Specific Quality of Life Outcomes after Endoscopic Endonasal Surgery of the Sella.

OBJECTIVE: To assess olfactory outcomes as measured by an olfactory-specific quality of life (QOL) questionnaire in patients undergoing EESBS for sellar lesions. DESIGN: Retrospective case series. SETTING: Tertiary academic medical center. PARTICIPANTS: In total, 36 patients undergoing EESBS for lesions limited to the sella were evaluated. MAIN OUTCOME MEASURES: The following were performed before and three months after surgery: 22-Item Sinonasal Outcomes Test (SNOT-22), University of Pennsylvania Smell Identification Test (UPSIT), and the Assessment of Self-reported Olfactory Functioning (ASOF), which has three domains: subjective olfactory capability scale (SOC), smell-related problems (SRP), and olfactory-related quality of life (ORQ). RESULTS: Median age at surgery was 52.5 years, with a median tumor size of 1.8 cm (range: 0.2 to 3.9 cm). Pre- and postoperative median scores were 35 [34, 36.2] and 34.5 [32, 36] for UPSIT, 21 [7.5, 33.5] and 21.5 [6.8, 35.7] for SNOT-22, 10 [9, 10] and 9 [8, 10] for ASOF-SOC, 5 [4.8, 5] and 4.5 [4, 5] for ASOF-SRP, and 5 [5, 5] and 5 [4.5, 5] for ASOF-ORQ. There was no significant change in the two of the three domains of the ASOF. Correlation between ASOF and UPSIT scores were weak. Older age and larger tumor size were associated with worsened olfaction after surgery. CONCLUSIONS: Patients did not experience significant changes in olfactory-specific QOL three months after EESBS, as measured by two domains of the ASOF. The ASOF may serve as a useful adjunctive tool for assessing olfaction after surgery. The lack of correlation between UPSIT and ASOF suggests the need for more research in subjective olfactory-related quality of life after surgery.

Sinus Irrigation Penetration After Proposed Modified Draf IIa Technique in a Side-to-Side Cadaveric Model.

BACKGROUND: Endoscopic sinus surgery (ESS) is an effective intervention for patients with medically refractory chronic rhinosinusitis. Frontal sinusotomy is the most challenging part of ESS, with one of the key outcomes being access for topical irrigations. OBJECTIVE: The purpose of this study is to compare irrigation penetration into the frontal sinus following Draf IIa versus modified Draf IIa frontal sinusotomy. METHODS: Four fresh cadaver heads were used in this experiment. Draf IIa was performed on one side of each head and a modified Draf IIa on the contralateral side. This proposed modification consists of a Draf IIa combined with an agger nasi punch-out procedure and partial trimming of the vertical lamella of the middle turbinate back to the posterior table of the frontal sinus without drilling the beak. Each head was irrigated with methylene blue-dyed water and recorded by rigid endoscopy through an endonasal view (EV) of the frontal sinus and frontal trephination view (TV). Two blinded rhinologists scored the extent of staining (using an ordinal scale of 0 to 3) for each side. A case report where the modified Draf IIa was performed is also described. RESULTS: After modified Draf IIa sinuosotomy, the mean score for the EV was 2.125 and for the TV was 2, versus 0.875 and 0.625 for traditional Draf IIa, respectively. There was a statistically significant increase for both EV (p = 0.019) and TV (p = 0.018) after modified Draf IIa. CONCLUSION: In our cadaveric model, this procedural modification improved penetration of postoperative irrigations into the frontal sinus. This simple technique may be easily adapted into frontal ESS when indicated.

Sleep medicine in otolaryngology units: an international survey.

PURPOSE: No study to date has described the overall landscape of sleep disorders management and training in otolaryngology departments of different countries. The aim of our study was to investigate and compare settings, diagnostic and therapeutic approaches and training programmes. METHODS: An international online survey was developed with the collaboration of the YO-IFOS (Young Otolaryngologists-International Federation of Otorhinolaryngological Societies) to assess the current practice of otolaryngologists in the management of sleep disorders. The survey also included a session dedicated to training. RESULTS: A total of 126 otolaryngologists completed the survey. The larger part of responses was collected from Central/South America and Europe. The majority of responders from South/Central America (97%) declared to be certified as sleep specialist while 49% of Europeans stated the opposite. Of responders 83% perform a drug-induced sleep endoscopy (DISE) before planning a possible surgical intervention. Soft palate and base of tongue interventions were the most common procedure, respectively performed in 94% and 79% of the cases. Residents were allowed to perform soft palate surgery in 77% of the cases. Upper airway stimulation (26% vs 10%), trans-oral robotic surgery (36% vs 11%) and radiofrequency of the base of the tongue (58% vs 25%) were preferred more frequently by European responders. The highest caseloads of soft palate surgery and bi-maxillary advancement were registered in the academic institutions. CONCLUSION: Significant concordance and few interesting divergences in diagnosis and treatment of sleep disorders were observed between nationalities and types of institution. Economic resources might have played a significant role in the therapeutic choice. Trainees' lack of exposure to certain interventions and to a sufficient caseload appeared to be the main burden to overcome.

Causative Factors for Complications in Transpalatal Advancement.

INTRODUCTION: Transpalatal advancement (TPA) is a procedure that is used when modern variants of uvulopharyn-gopalatoplasty are unable to provide enough anterior traction. Although successful in reduction of obstructive sleep apnea (OSA) parameters, it also comes with procedure-specific risks. Formation of an oro-nasal fistula (ONF) is a complication that results in significant morbidity and a protracted treatment course. METHODS: After approval from the University of Wollongong Health Research Ethics Committee, a retrospective chart review of all cases undergoing TPA performed by a single surgeon over a 10-year period from 2008 to 2018 was performed. Patients underwent pre- and postoperative level 1 or 2 polysomnography. Factors potentially contributing to palatal complications, as well as pre- and postoperative polysomnographic parameters, subjective sleep questionnaires, and body mass index (BMI) were statistically analyzed where a P value <.05 was considered a significant result. RESULTS: A total of 59 patients were included. Overall palatal complication rate was 25.4% (15/59), with the most common being transient velo-palatal insufficiency (VPI) (8/59, 13.6%). ONF developed in 4/59 (6.8%) of patients. None of the analyzed contributing factors for palatal complications were statistically significant, except the presence of a high-arched palate and development of ONF. All analyzed sleep parameters, as well as BMI, were significantly different when comparing pre- to postoperative results. CONCLUSION: This study suggests that TPA has a role in current sleep surgery paradigms and can significantly improve both objective and subjective outcome measures of OSA. Surgeons contemplating TPA on patients with high-arched hard palates should do so with caution.

Focus Quality Assessment of High-Throughput Whole Slide Imaging in Digital Pathology.

One of the challenges facing the adoption of digital pathology workflows for clinical use is the need for automated quality control. As the scanners sometimes determine focus inaccurately, the resultant image blur deteriorates the scanned slide to the point of being unusable. Also, the scanned slide images tend to be extremely large when scanned at greater or equal 20X image resolution. Hence, for digital pathology to be clinically useful, it is necessary to use computational tools to quickly and accurately quantify the image focus quality and determine whether an image needs to be re-scanned. We propose a no-reference focus quality assessment metric specifically for digital pathology images that operate by using a sum of even-derivative filter bases to synthesize a human visual system-like kernel, which is modeled as the inverse of the lens' point spread function. This kernel is then applied to a digital pathology image to modify high-frequency image information deteriorated by the scanner's optics and quantify the focus quality at the patch level. We show in several experiments that our method correlates better with ground-truth z -level data than other methods, which is more computationally efficient. We also extend our method to generate a local slide-level focus quality heatmap, which can be used for automated slide quality control, and demonstrate the utility of our method for clinical scan quality control by comparison with subjective slide quality scores.

Symptom Profile of Chronic Rhinosinusitis Versus Obstructive Sleep Apnea in a Tertiary Rhinology Clinic.

OBJECTIVES: Patients with undiagnosed obstructive sleep apnea (OSA) often present to an otolaryngologist with symptoms of chronic rhinosinusitis (CRS). This study analyzes symptom profiles of such patients to help identify those who may benefit from polysomnography. METHODS: This is a 2-year retrospective analysis of adult patients at an academic practice with a rhinologic chief complaint. Results of the 22-Item Sinonasal Outcomes Test (SNOT-22) were compared between patients with untreated OSA without CRS (OSA group) and a control group of CRS patients (CRS group). Univariate analysis with Bonferroni correction and multiple logistic regression were used. RESULTS: Forty-one patients were included in the OSA group (mean apnea-hypopnea index = 28.8, SD = 21.0) and 124 patients in the CRS group. CRS group demonstrated higher scores in nasal, extra-nasal, and ear/facial symptom domains (P < .001), while OSA group displayed higher psychological (P = .020) and sleep domain scores (P = .048). Nasal discharge (P < .001) and loss of smell/taste (P = .004) scores were higher in CRS group, whereas facial pain (P = .285) and nasal obstruction (P = .822) were not significantly different. On logistic regression, thick nasal discharge, loss of smell/taste, and ear pain were predictive of CRS, while waking up tired was predictive of OSA. Finally, OSA group was more likely to choose a sleep-related issue as a most important symptom (MIS) (P < .001). CONCLUSIONS: OSA and CRS have significant overlap in symptom profiles. The SNOT-22 can help identify those with undiagnosed OSA. OSA should be suspected in patients who report a sleep-related item as a MIS and display higher psychological and sleep domain scores.

Idiopathic pretracheal deep neck space infection with mediastinal extension: A series of 3 cases and review of the literature.

Idiopathic pretracheal deep neck space infection is an extremely rare condition with potentially devastating complications. We present a series of 3 cases of pretracheal deep neck space infection that arose in the absence of trauma or a congenital lesion and that exhibited mediastinal spread. To the best of our knowledge, these cases represent the first reported series of this rare condition to be published in the English-language literature. All cultures grew Streptococcus milleri, and all patients had a favorable outcome. A high index of suspicion for a deep neck space infection is warranted in view of the devastating complications of this condition. Computed tomography is the investigation of choice. Treatment with intravenous antibiotics and surgical drainage, particularly when mediastinitis is present, is recommended. This rare presentation warrants a thorough investigation to identify the source of infection.

Atypical Pharyngeal Pouch Arising Bilaterally between the Hyoid Bone and Thyroid Cartilage.

Introduction. Pharyngoesophageal diverticuli are a common cause of dysphagia; they are associated with various morbidities and a decreased quality of life. There are several different types of the diverticuli, and they are divided based on the anatomical location of origin relative to the cricopharyngeal muscle; these include Zenker's, Killian-Jamieson's, and Laimer's diverticula. The authors present a unique case of pharyngoesophageal diverticulum that has not been previously described. Case Presentation. A 65-year-old male presented with a 12-month history of dysphagia and odynophagia for solids. Barium swallow revealed bilateral moderately sized diverticuli that altered in size during the different phases of swallow. CT scan of the neck with oral contrast further identified the anatomy of the diverticuli, arising between the hyoid bone and thyroid cartilage. Discussion. An external transcervical approach was utilised to successfully repair the diverticuli. Subsequent cricopharyngeal spasm was treated with botulinum toxin, and the patient recovered with no ongoing symptoms. The barium swallow study is a commonly utilised initial investigation as it is easy to perform and safe and has good diagnostic value. Definitive management usually involves either endoscopic or open surgery. This case depicts a unique case of a pharyngeal diverticulum arising between the hyoid bone and thyroid cartilage.

Surgical Approaches to Obstructive Sleep Apnea.

Surgery in adult obstructive sleep apnea (OSA) has undergone significant advancement in recent years and continues to evolve. It is a modality of treatment used in the context of failed device use, specifically, failed continuous positive airway pressure or mandibular advancement splint. In this context, the role of surgery is either as salvage therapy or to facilitate better tolerance of device use. Other treatments such as weight loss, adjuvant nasal therapy (medical ± prephase nasal surgery) and positional devices may be combined with airway surgery. In general, patients with OSA are managed with in-hospital monitoring perioperatively.

Frontal sinus surgery and sinus distribution of nasal irrigation.

BACKGROUND: Effective mucus lavage and delivery of topical pharmaceuticals are central to successful management of chronic rhinosinusitis (CRS). The frontal sinus remains difficult to penetrate with topical therapies. This study evaluates the benefit of Draf III frontal dissection compared to traditional Draf IIa for distribution of topical therapies. METHODS: Fresh human cadaver heads were dissected sequentially with Draf IIa frontal sinusotomy and then Draf III procedures. Each cavity was irrigated with pediatric (120 mL) and adult (240 mL) irrigation bottles with 1/1000 10% fluorescein-labeled free water in 2 fixed positions (vertex and Frankfort horizontal). An endoscope at a fixed position within the frontal sinus recorded frontal sinus and frontal recess penetration. The images then underwent blinded evaluation of fluid distribution scored as 0 to 4 (nasal cavity only, frontal recess, medial one-half, lateral one-half, and lavage). Ordinal distribution score was analyzed with Kendall's tau-b. RESULTS: Eight specimens (age 76 ± 11.2 years; 50% female) were assessed. Draf III was superior to Draf IIa in ability to achieve frontal sinus distribution of irrigation (90.6% vs 50.1%, p < 0.001). Vertex head position improved distribution (90.6% vs 50.1%, p < 0.001), was synergistic with Draf III (100% with 87.5% lavage, p < 0.001), but was unable to overcome Draf IIa (81.2% with 25% lavage, p < 0.001). Irrigation volume trended toward improved distribution with larger volume irrigations. CONCLUSION: Successful treatment of sinonasal disease may require postoperative delivery of topical therapies. Draf III frontal sinusotomy achieves superior topical access, and access to the frontal sinus with Draf IIa appears limited, despite large volumes and positioning.

Laryngotracheal stenosis and airway surgery--an outcomes based approach.

OBJECTIVES: To review the outcomes of endoscopic, open or a combination of both surgical modalities for laryngotracheal stenosis and establish which factors influence results. METHODS: Records of all children undergoing laryngotracheal procedures (excluding laryngomalacia and aspirated foreign bodies) by the Department of Otolaryngology at The Children's Hospital at Westmead between January 2003 and November 2011 were reviewed. Specific data on population, intervention, covariates and outcomes were recorded and analysed. RESULTS: A total of 104 patients undergoing 277 procedures were included. 211 (76%) of the procedures were endoscopic, remaining 66 (24%) open. Patients undergoing open surgery were more likely to have significant co-morbidity, prior intubation, require ICU admission or tracheostomy and have a longer hospital stay. 57 (54.8%) patients were successfully treated with a single procedure (48 endoscopic and 9 open). Of the endoscopic patients requiring further surgery, 16 were managed with multiple endoscopic procedures, whilst 12 underwent subsequent open procedures. Open surgery was performed on 66 patients, 63.6% (42/66) of all open procedures required further endoscopic intervention and 45.2% (19/42) of these avoided further open surgery. CONCLUSIONS: Both open and endoscopic surgery have a role in laryngotracheal stenosis, and many patients benefit from a combination of both. Ultimately the decision depends on experience of the treating team, social considerations, and institutional capabilities. A multi-centre prospective data collection would be a useful tool to further investigate optimal management approach.

Skin excisions: not so simple for the regionally based general surgical trainee.

BACKGROUND: In the regional setting, general practitioners often refer complicated skin excisions to general surgery, whereas in larger centres this is the domain of plastic surgery. General surgical trainees often do not have adequate exposure to complex skin excisions prior to placement in regional centres. OBJECTIVE: To explore what factors affected positive margin rates in surgical registrars in a regional setting. DESIGN: Retrospective audit. SETTING: Large teaching hospital (referral centre). PARTICIPANTS: All skin lesions excised under local anaesthetic by registrars at a single referral centre over a 30-month period from 2007 to mid 2009, of these only basal cell carcinomas (BCC) and squamous cell carcinomas (SCC) were analysed. RESULTS: Registrars excised a total of 703 skin lesions, 314 (43.4%) were BCC or SCC, and of these 50 (15.9%) had positive margins. Repeated measures multivariable logistic regression was performed on relevant data. Surgical registrars had significantly higher positive margin rates when operating unsupervised (P=0.014). Although not significant, there was a tendency for BCC excisions to have positive margins (P=0.059). There was no statistical difference when comparing lesions excised on the head compared to body, use of a graft/flap or registrar training level. CONCLUSION: This study's positive margin rate of 15.9% falls within the range reported in the literature (0.7% to 20.7%); however, this has the potential to be further reduced. Surgical registrars excising skin lesions in regional centres, regardless of level of training, should have closer supervision. Regular surgical audit should be done so registrars can have early feedback on performance.