Low but measurable stimulated serum thyroglobulin levels <2 µg/L frequently predict incomplete response in differentiated thyroid cancer patients.

INTRODUCTION: The study was aimed to determine the response and predictive risk factors of differentiated thyroid cancer (DTC) with measurable (0.4-2.0 µg/L) stimulated serum thyroglobulin (sTg) during the 10-24 months after radioiodine remnant ablation (RRA) and their long-term outcomes. METHODS: Out of 839 retrospectively reviewed patients, 95 eligible DTC patients were included. Patients were classified as having incomplete response or no evidence of disease (NED). The sTg cut-off values with highest predicted accuracy for incomplete response at 10-24 months were calculated with receiver operator characteristics curve analysis. RESULTS AND CONCLUSION: At 10-24 months after RRA, incomplete response was identified in 54 patients (57%) and 38/54 (70.4%) patients were found with structural evidence of disease. The remaining 16 patients (29.6%) had biochemical evidence of disease without structural evidence of disease. Forty-one patients (43%) were classified as having NED at 10-24 months after RRA and 27 patients (66%) did not receive further radioactive iodine (RAI) therapy and remained disease free at median follow-up of 6.5 years. Fourteen patients received second RAI treatment after 6 months and before the 10-24 months assessment time point. Of these, 2 had persistent tumor 6 years later. The sTg >0.6 µg/L at 6-10 months after RRA had optimal sensitivity (83.3%), specificity (56%) and negative predictive value (72%) of detecting incomplete response at 10-24 months after RRA. A total of 23/43 patients in the American Thyroid Association low-risk category had incomplete response after first RRA and 5/23 (21.7%) had recurrent/persistent disease at long-term follow-up.

Building a Bridge to Community: A Pragmatic Randomized Trial Examining a Combined Physical Therapy and Resistance Exercise Intervention for People after Head and Neck Cancer.

BACKGROUND: Established barriers to general exercise and physical activity among individuals with head and neck cancer include dry mouth, difficulty eating, weight loss, fear of injury, comorbidities, and treatment-related symptoms of pain and fatigue. METHODS/DESIGN: A 12-week pragmatic randomized controlled trial was conducted followed by an optional supported exercise transition phase. Eligible participants were individuals with head and neck cancers who had undergone surgery and/or radiation therapy to lymph node regions in the neck. Participants were randomized to a comparison group involving a shoulder and neck physiotherapeutic exercise protocol, or to a combined experimental group comprising the shoulder and neck physiotherapeutic exercise protocol and lower-body resistance exercise training. The primary outcome of this study was fatigue-related quality of life. RESULTS: Sixty-one participants enrolled, 59 (97%) completed the randomized trial phase, 55 (90%) completed the 24-week follow-up, and 52 (85%) completed the one-year follow-up. Statistically significant between-group differences were found in favor of the combined experimental group for the fatigue-related quality of life, fitness outcomes, and overall physical activity. Paired comparisons confirmed significant within-group improvements for both groups from baseline to one-year follow-up across most outcomes. DISCUSSION: A group-based combined physiotherapeutic and lower-body resistance exercise program was feasible and effective. Findings are limited to individuals who had undergone a surgical neck dissection procedure. Given the complexity of head and neck cancer, further pragmatic interdisciplinary research is warranted.

Quantifying the severity of sarcopenia in patients with cancer of the head and neck.

BACKGROUND & AIMS: Existing skeletal muscle index (SMI) thresholds for sarcopenia are inconsistent, and do not reflect severity of depletion. In this study we aimed to define criterion values for moderate and severe skeletal muscle depletion based on the risk of mortality in a population of patients with head and neck cancer (HNC). Additionally, we aimed to identify clinical and demographic predictors of skeletal muscle depletion, evaluate the survival impact of skeletal muscle depletion in patients with minimal nutritional risk or good performance status, and finally, benchmarking SMI values of patients with HNC against healthy young adults. METHODS: Population cohort of 1231 consecutive patients and external validation cohorts with HNC had lumbar SMI measured by cross-sectional imaging. Optimal stratification determined sex-specific thresholds for 2-levels of SMI depletion (Class I and II) based on overall survival (OS). Adjusted multivariable regression analyses (tumor site, stage, performance status, age, sex, dietary intake, weight loss) determined relationships between 2-levels of SMI depletion and OS. RESULTS: Mean SMI (cm2/m2) was 51.7 ± 9.9 (males) and 39.8 ± 7.1 (females). The overall and sex-specific population demonstrated an increased risk of mortality associated with decreasing SMI. Sex-specific SMI (cm2/m2) depletion thresholds for 2-levels of muscle depletion determined by optimal stratification for males and females, respectively (male: 45.2-37.5, and <37.5; female: 40.9-34.2, and <34.2). In the overall population, Normal SMI, Class I and II SMI depletion occurred in 65.0%, 24.0%, and 11.0%, respectively. Median OS was: Normal SMI (114 months, 95% CI, 97.1-130.8); Class I SMI Depletion (42 months, 95% CI, 28.5-55.4), and Class II SMI Depletion (15 months, 95% CI, 9.8-20.1). Adjusted multivariable analysis compared with Normal SMI (reference), Class I SMI Depletion (HR, 1.49; 95% CI, 1.18-1.88; P < .001), Class II SMI Depletion (HR, 1.91; 95% CI, 1.42-2.58; P < .001). CONCLUSIONS: Moderate and severe SMI depletion demonstrate discrimination in OS in patients with HNC. Moderate and severe SMI depletion is prevalent in patients with minimal nutrition risk and good performance status. Benchmarking SMI values against healthy young adults exemplifies the magnitude of SMI depletion in patients with HNC and may be a useful method in standardizing SMI assessment.

Predictors of adherence to an exercise program for shoulder pain and dysfunction in head and neck cancer survivors.

PURPOSE: Achieving acceptable levels of adherence to exercise may be a challenge with head and neck cancer (HNC) survivors given the high morbidity associated with treatment. The purpose of the present trial was to identify the key predictors of adherence during our shoulder exercise rehabilitation trial. METHODS: Fifty-two HNC survivors were randomly assigned to a 12-week progressive resistance exercise training protocol (n = 27) or a standardized therapeutic exercise protocol (n = 25) that was prescribed to address shoulder pain and dysfunction. Baseline data were collected on standard demographic, medical, behavioral, symptom, psychosocial, and motivational variables from the theory of planned behavior. RESULTS: The exercise adherence rate for the trial was 91%. In multivariate analysis, the independent predictors of reduced adherence were undergoing a more extensive neck dissection procedure (ß = -0.361; P = 0.007) and reporting daily alcohol consumption (ß = -0.298; P = 0.031). Higher exercise adherence was achieved by HNC participants who had undergone nerve sparing neck dissection procedures and who were not regular drinkers. CONCLUSION: Excellent adherence to exercise was achieved in the trial despite high morbidity associated with HNC treatment. The high adherence achieved was likely due to the select and highly motivated sample of HNC survivors as well as to factors associated with trial design such as the support offered to participants. The findings of this trial need to be further explored and confirmed in a larger study that includes a more diverse sample of HNC survivors.

Patterns and predictors of relapse in Merkel cell carcinoma: Results from a population-based study.

BACKGROUND AND PURPOSE: Prospective data evaluating the role of adjuvant radiotherapy (RT) for Merkel Cell Carcinoma(MCC) is lacking. To better understand the efficacy of adjuvant RT, a population-based patterns of failure study was conducted. METHODS: We identified MCC patients treated from 1988 to 2018.Primary outcome measures were recurrence-free survival (RFS), overall survival (OS) and MCC-specific survival (MCC-SS). Charlson Co-morbidity Index (CCI) was also calculated. RESULTS: 217 patients with mean age 79 (range: 33-96) were analyzed. The median follow-up was 40 months. Treatments were: surgery(S) alone (n = 101, 45%) or S + RT(n = 116, 55%).Local recurrence (LR) was low in stage I (n = 6, 6.5%) with clear margin of ≥1 cm, negative sentinel lymph node biopsy (SLNB) without high-risk factors, irrespective of adjuvant RT. Tumor size ≥ 2 cm (HR:2.95; p = 0.024) and immunosuppression(HR:3.98; p = 0.001) were associated with high risk of nodal failure. Adjuvant RT was associated with significant reduction in regional failure (HR:0.36; p = 0.002). Distant metastases (DM) were infrequent in stage I (4/90) and stage II (4/34), compared to stage III (32/93). Adjuvant RT improvedRFS but did not influence MCC-SS and OS. CCI was a significant predictor of OS. CONCLUSIONS: Adjuvant RT improvedRFS, withoutimpact on MCC-SS and OS. Co-morbidity rather than RT influenced OS. Adjuvant RT may be avoided instage I patients with negative SLNB and no associated high-risk factors. Prophylactic RNI could be considered in stage II with high risk features, inspite of negative SLNB. Stage III patients benefited from adjuvant RNI, but no impact on prevention of DM.

An efficient CNN based algorithm for detecting melanoma cancer regions in H&E-stained images.

Histopathological images are widely used to diagnose diseases such as skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of abnormal cell nuclei and their distribution within multiple tissue sections would enable rapid comprehensive diagnostic assessment. In this paper, we propose a deep learning-based technique to segment the melanoma regions in Hematoxylin and Eosin-stained histopathological images. In this technique, the nuclei in an image are first segmented using a deep learning neural network. The segmented nuclei are then used to generate the melanoma region masks. Experimental results show that the proposed method can provide nuclei segmentation accuracy of around 90% and the melanoma region segmentation accuracy of around 98%. The proposed technique also has a low computational complexity.

Automated proliferation index calculation for skin melanoma biopsy images using machine learning.

The Proliferation Index (PI) is an important diagnostic, predictive and prognostic parameter used for evaluating different types of cancer. This paper presents an automated technique to measure the PI values for skin melanoma images using machine learning algorithms. The proposed technique first analyzes a Mart-1 stained histology image and generates a region of interest (ROI) mask for the tumor. The ROI mask is then used to locate the tumor regions in the corresponding Ki-67 stained image. The nuclei in the Ki-67 ROI are then segmented and classified using a Convolutional Neural Network (CNN), and the PI value is calculated based on the number of the active and the passive nuclei. Experimental results show that the proposed technique can robustly segment (with 94 % accuracy) and classify the nuclei with a low computational complexity and the calculated PI values have less than 4 % average error.

Detection of malignant melanoma in H&E-stained images using deep learning techniques.

Histopathological images are widely used to diagnose diseases including skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of all abnormal cell nuclei and their distribution within multiple tissue sections would assist rapid comprehensive diagnostic assessment. In this paper, we propose a deep learning-based technique to segment the melanoma regions in Hematoxylin and Eosin (H&E) stained histopathological images. In this technique, the nuclei in the image are first segmented using a Convolutional Neural Network (CNN). The segmented nuclei are then used to generate melanoma region masks. Experimental results with a small melanoma dataset show that the proposed method can potentially segment the nuclei with more than 94 % accuracy and segment the melanoma regions with a Dice coefficient of around 85 %. The proposed technique also has a small execution time making it suitable for clinical diagnosis with a fast turnaround time.

Dentition, nutritional status and adequacy of dietary intake in treatment naïve head and neck cancer patients.

OBJECTIVES: To examine the relationship of reduced numbers of occluding teeth and dietary intake (DI), nutrition impact symptoms (NIS), and weight loss (WL) in head and neck cancer (HNC) patients. METHODS: As a part of the standard of care, treatment-naïve HNC patients (n = 104) completed dental evaluation (number of teeth, total anterior/posterior occlusal teeth, Eichner Index (EI) classification), WL, DI questionnaire and HNC Symptom Checklist©. Descriptive statistics (Kruskal-Wallis, Fisher-exact, χ2 tests) and (uni-) multi-variable logistic regression. RESULTS: Overall, 42, 45 and 13% of patients were in EI-class A, B and C with a median of 8, 3, and 0 total posterior occlusal teeth. EI-class B/C patients were older, more likely to have impaired DI (OR = 3.88; 95%CI:1.63-9.26; P = 0.002) and reported interference with DI by 11 NIS (p < 0.05). DI was, however, reported as unimpaired in 77, 49 and 39% of patients in EI-class A, B and C, respectively. The subset of EI-class B/C patients with impaired DI, had more NIS interference with DI (P < 0.05; difficulty chewing, pain, early satiety, lack of energy); EI-class C patients additionally had dry mouth, thick saliva and dysphagia (P < 0.05). In logistic regression, EI-classes B/C patients with reduced (vs unimpaired) DI were more likely to have ≥5% WL (OR = 10.1; 95%CI:2.0-50.0), higher NIS interference (range OR 4.3-10.7). CONCLUSIONS: More than half of these HNC patients had reduced numbers of occlusal teeth or were edentulous. EI-class B/C patients did not necessarily have impaired DI, however the combination of EI-class B/C and a constellation of NIS, associated with reduced DI. CLINICAL SIGNIFICANCE: Treatment naïve head and neck cancer (HNC) patients with reduced occlusal and masticatory performance (Eichner Index B/C) and reduced dietary intake are at high risk for weight loss. Identifying HNC patients at risk may improve their oral health, dietary intake and reduce their risk of weight loss.

Prevalence and prognostic significance of malnutrition in patients with cancers of the head and neck.

BACKGROUND & AIMS: Head and neck cancer (HNC) are at high nutritional risk; however the prevalence and severity of malnutrition in contemporary patients with HNC are unclear. Diagnostic criteria for cancer-associated weight loss grading (WLG) (Grades 0 to 4) [1] have been validated and are recommended in oncology nutrition clinical practice guidelines [2-3]. The aim was to determine the prevalence using WLG in HNC patients and determine the extent to which reduced dietary intake (DI) explained variation of WLG. METHODS: A population-based cohort of HNC patients (N = 1756) in northern Alberta, Canada included consecutive new patients, 2004-2016. At referral to the regional cancer center weight history and DI categories were collected. Multinomial logistic regression (MLR) identified predictors of weight loss (WL) severity. Overall survival (OS) in relation to WL Grade and DI was determined by multivariable Cox proportional hazard. RESULTS: WL was absent in 42.9% and the remainder had Grade 1 (18%), Grade 2 (14.7%), Grade 3 (15.9%) and Grade 4 (8.5%) WL. Independent predictors of WLG in adjusted MLR model, included stage (P < 0.000), performance status (PS) (P < 0.000) and DI categories (P < 0.000); sex, age and disease site were not significant. Compared to "normal food in normal amount" adjusted Odds Ratio for WL Grade 4 was 4.0 (2.1-7.5) "normal food, but less than normal amount"; 25.2 (10.7-59.1) "little solid food"; 51.8 (10.5-255.3) "very little of anything"; 42.4 (11.0-163.0) "only liquids"; 25.9 (7.1-94.3) "only nutritional supplements". In the Cox model controlled for age, sex, cancer stage and site and PS, both WLG ((P < 0.000) and DI categories (P = 0.003) independently predicted OS. CONCLUSION: Data from this population cohort provide a benchmark for prevalence of cancer associated WL severity at diagnosis. Patient reported DI categories are strong predictors of WL and prognostic for OS.

Functional outcomes of the modified submandibular gland transfer procedure.

OBJECTIVES: Treatment for advanced head and neck cancers typically includes surgery followed by radiation therapy (RT). Radiation-induced xerostomia is a common sequela of these treatments. The modified submandibular gland transfer (M-SGT) procedure was developed to decrease xerostomia in the treatment of oral cavity cancer by sparing one submandibular gland (SMG) from radiation. This study's objectives were to: 1) elucidate the radiation-sparing capacity of the M-SGT, and 2) study the xerostomia-reducing potential of the M-SGT based on the University of Washington Quality-of-Life Questionnaire (UW-QOL). METHODS: Radiation therapy treatment plans were reviewed for all patients treated with surgery and RT who had a M-SGT at the University of Alberta Hospital during the study period. Outcomes included: 1) radiation dose received by the transferred SMG within the periparotid area compared to the submandibular triangle (ST), and 2) patient-reported saliva scores on the UW-QOL compared to historical controls without a gland transfer. RESULTS: Twenty-two patients were included. The mean radiation dose received by the transferred SMG was 29.00 grays (Gy) (standard deviation 14.59 Gy), thus reducing the mean radiation dose to the SMG by a statistically significant 18.34 Gy (confidence interval 95% (13.37, 23.32), P < 0.01) compared to the ST and below the D50 of the SMG (34 Gy). Sixty-five percent of patients rated their saliva as normal or mildly reduced on the UW-QOL as compared to 16% of controls (P = 0.01). CONCLUSION: The M-SGT technique is successful at reducing the radiation dose sustained by the SMG during adjuvant treatment and provides a significant improvement in xerostomia-related functional outcomes as compared to historical controls not receiving a gland transfer. LEVEL OF EVIDENCE: 4 Laryngoscope, 130:925-929, 2020.

Understanding the human salivary metabolome.

Saliva is a readily accessible biofluid that is important for the overall health, aiding in the chewing, swallowing, and tasting of food as well as the regulation mouth flora. As a first step to determining and understanding the human saliva metabolome, we have measured salivary metabolite concentrations under a variety of conditions in a healthy population with reasonably good oral hygiene. Using (1)H NMR spectroscopy, metabolite concentrations were measured in resting (basal) and stimulated saliva from the same subject and compared in a cohort of healthy male non-smoking subjects (n = 62). Almost all metabolites were higher in the unstimulated saliva when compared to the stimulated saliva. Comparison of the salivary metabolite profile of male smokers and non-smokers (n = 46) revealed citrate, lactate, pyruvate, and sucrose to be higher and formate to be lower in concentration in smokers compared with non-smokers (p < 0.05). Gender differences were also investigated (n = 40), and acetate, formate, glycine, lactate, methanol, propionate, propylene glycol, pyruvate, succinate, and taurine were significantly higher in concentration in male saliva compared to female saliva (p < 0.05). These results show that differences between male and female, stimulated and unstimulated, as well as smoking status may be observed in the salivary metabolome.

Novel lymph node segmentation and proliferation index measurement for skin melanoma biopsy images.

The lymphatic system is the immune system of the human body, and includes networks of vessels spread over the body, lymph nodes, and lymph fluid. The lymph nodes are considered as purification units that collect the lymph fluid from the lymph vessels. Since the lymph nodes collect the cancer cells that escape from a malignant tumor and try to spread to the rest of the body, the lymph node analysis is important for staging many types skin and breast cancers. In this paper, we propose a Computer Aided Diagnosis (CAD) method that segments the lymph nodes and melanoma regions in a biopsy image and measure the proliferation index. The proposed method contains two stages. First, an automated technique is used to segment the lymph nodes in a biopsy image based on histogram and high frequency features. In the second stage, the proliferation index for the melanoma regions is calculated by comparing the number of active and passive nuclei. Experimental results on 76 different lymph node images show that the proposed segmentation technique can robustly segment the lymph nodes with more than 90% accuracy. The proposed proliferation index calculation has low complexity and has an average error rate of less than 1.5%.

Cetuximab-radiotherapy combination in the management of locally advanced cutaneous squamous cell carcinoma.

INTRODUCTION: We report the outcomes of using a combination of cetuximab with radiation therapy (Cetux-RT) to treat a selected group of patients with locally advanced (unresectable) cutaneous squamous cell carcinoma (LA-cSCC). This study presents two-year efficacy and safety data for 8 patients with LA-cSCC treated within a single institution. METHODS: Between 2014 and 2017 a total of eight patients (seven males, one female) with LA-cSCC received curative intent treatment with Cetux-RT. All patients received an initial loading dose of cetuximab at 400 mg/m2 seven days prior to radiotherapy, followed by weekly treatment with 250 mg/m2 , continuing through the end of radiotherapy. Radiation doses were 6600 cGy/30 fr (n = 2), 6300 cGy/30fr (n = 2) and 5500 cGy/22 fr (n = 4). RESULTS: The median age was 81 years (range, 55-87). The ECOG performance status of all patients was between 0 and 2. With a median duration of follow-up of 25 months (range 10-48 months), five patients remain in a complete response. After a partial response, another patient has relapsed and is receiving palliative chemotherapy, while two patients have died during the period of follow up (one of whom died following progression of disease, the other of an unrelated cause). Treatment in this group of patients was well tolerated, with most toxicities ≤ grade 2, and no toxicities of grade 4/5 reported. CONCLUSIONS: Cetux-RT was well tolerated and provided durable disease control within this patient sample. Our data support the use of the Cetux-RT regimen for selected patients with inoperable LA-cSCC and adequate performance status.

Automated analysis and classification of melanocytic tumor on skin whole slide images.

This paper presents a computer-aided technique for automated analysis and classification of melanocytic tumor on skin whole slide biopsy images. The proposed technique consists of four main modules. First, skin epidermis and dermis regions are segmented by a multi-resolution framework. Next, epidermis analysis is performed, where a set of epidermis features reflecting nuclear morphologies and spatial distributions is computed. In parallel with epidermis analysis, dermis analysis is also performed, where dermal cell nuclei are segmented and a set of textural and cytological features are computed. Finally, the skin melanocytic image is classified into different categories such as melanoma, nevus or normal tissue by using a multi-class support vector machine (mSVM) with extracted epidermis and dermis features. Experimental results on 66 skin whole slide images indicate that the proposed technique achieves more than 95% classification accuracy, which suggests that the technique has the potential to be used for assisting pathologists on skin biopsy image analysis and classification.

Combination of submandibular salivary gland transfer and intensity-modulated radiotherapy to reduce dryness of mouth (xerostomia) in patients with head and neck cancer.

BACKGROUND: Xerostomia is a debilitating side effect of radiotherapy for head and neck cancer. Combining surgical submandibular-gland transfer (SMGT) with intensity-modulated radiotherapy (IMRT) may provide greater protection of salivary function. METHODS: This was a single-institution, prospective phase II feasibility trial. Patients with head and neck cancer or unknown primary with neck node metastases received primary surgery with SMGT and postoperative radiotherapy with tomotherapy (60 Gy in 30 fractions). Toxicity and quality of life (QOL) were assessed before surgery, before RT, and after RT. RESULTS: Forty patients received SMGT and IMRT. Only 1 patient experienced grade 3 salivary gland toxicity. At 12 months post-RT, the rate of absent or only mild xerostomia was 89%, and salivary flow rates were approximately 75% of pre-RT levels. CONCLUSIONS: The combination of IMRT with SMGT is feasible and with improved dose constraints may maximally spare the parotid and submandibular glands, leading to decreased xerostomia and improved patient QOL.

Sparing the parotid glands and surgically transferred submandibular gland with helical tomotherapy in post-operative radiation of head and neck cancer: a planning study.

BACKGROUND AND PURPOSE: To evaluate the feasibility of sparing the parotid glands and surgically transferred submandibular gland (SMG) by intensity modulated radiotherapy (IMRT) in post-operative cases of head and neck cancer (HNC). MATERIALS AND METHODS: Ten patients (larynx-2, base of tongue-4, tonsil-3, and unknown primary-1; pathologic stages III-IV) who underwent SMG transfers on the side of N0 neck along with definitive surgery were selected for this study. IMRT planning was done retrospectively using helical tomotherapy approach. Planning objective was to deliver 60 Gy to PTV1 and 54 Gy to PTV2 while maintaining the mean dose to the total parotid volume (TPV) and SMG less than 26 Gy. RESULTS: The mean dose (+/-SD) to the TPV and SMG were 25+/-0.6 Gy and 23+/-1.9 Gy, respectively. The D(95) for PTV1 and PTV2 were 59.9+/-0.1 Gy and 54.9+/-0.3 Gy, respectively, satisfying our planning goal for PTV coverage. The D(99) for PTV1 and PTV2 were 58.2+/-0.7 Gy and 49.5+/-2.2 Gy, respectively, showing that sparing the salivary glands did not result in underdosing of the PTVs. CONCLUSIONS: By combining the gland transfer and IMRT, the mean dose to TPV and transferred SMG could be reduced to less than 26 Gy in post-operative patients of HNC.

Automatic Nuclei Detection Based on Generalized Laplacian of Gaussian Filters.

Efficient and accurate detection of cell nuclei is an important step toward automatic analysis in histopathology. In this work, we present an automatic technique based on generalized Laplacian of Gaussian (gLoG) filter for nuclei detection in digitized histological images. The proposed technique first generates a bank of gLoG kernels with different scales and orientations and then performs convolution between directional gLoG kernels and the candidate image to obtain a set of response maps. The local maxima of response maps are detected and clustered into different groups by mean-shift algorithm based on their geometrical closeness. The point which has the maximum response in each group is finally selected as the nucleus seed. Experimental results on two datasets show that the proposed technique provides a superior performance in nuclei detection compared to existing techniques.

Automatic measurement of melanoma depth of invasion in skin histopathological images.

Measurement of melanoma depth of invasion (DoI) in skin tissues is of great significance in grading the severity of skin disease and planning patient's treatment. However, accurate and automatic measurement of melanocytic tumor depth is a challenging problem mainly due to the difficulty of skin granular identification and melanoma detection. In this paper, we propose a technique for measuring melanoma DoI in microscopic images digitized from MART1 (i.e., meleanoma-associated antigen recognized by T cells) stained skin histopathological sections. The technique consists of four modules. First, skin melanoma areas are detected by combining color features with the Mahalanobis distance measure. Next, skin epidermis is segmented by a multi-thresholding method. The skin granular layer is then identified based on Bayesian classification of segmented skin epidermis pixels. Finally, the melanoma DoI is computed using a multi-resolution approach with Hausdorff distance measurement. Experimental results show that the proposed technique provides a superior performance in measuring the melanoma DoI than two closely related techniques.