11  Descriptive Statistics

This chapter provides a detailed descriptive analysis of the consultation dataset, covering pathologist activity, subspecialty distributions, consultation topics, and organ systems.

[1] TRUE

11.1 Pathologist Activity

11.1.1 Consultations Requested (Asker)

Top 15 Pathologists by Consultation Requests
Asker Asker_Sub Total_Requests
P8 GIS 754
P13 Uro 684
P10 Gyn 434
P4 GIS 345
P18 Lung 319
P2 BST 275
P28 Hemato 241
P3 General 228
P17 Skin 207
P21 Gyn 197
P9 GIS 194
P22 General 185
P5 Hemato 176
P11 Pancreas 171
P1 Gyn 158

11.1.2 Consultations Answered (Responder)

Top 15 Pathologists by Responses
Responder Responder_Sub Total_Responses
P5 Hemato 751
P9 GIS 696
P2 BST 684
P11 Pancreas 522
P23 GIS 407
P21 Gyn 399
P17 Skin 362
P8 GIS 348
P33 261
P6 GIS 254
P19 Breast 227
P10 Gyn 216
P28 Hemato 124
P24 BST 120
P27 Breast 94

11.2 Subspecialty Distribution

Distribution of consultation volume based on the subspecialty of the Asker.

11.3 Consultation Categories (13-Category Analysis)

Analysis of consultation topics using a 13-category classification system with multi-label tagging.

Distribution of Consultation Categories by Unique Cases
Category Cases Percentage (%)
Dysplasia/Grade 1022 22.6
Hematopathology 827 18.2
Other 449 9.9
Inflammatory/Non-neoplastic 447 9.9
Cytology/FNA 423 9.3
Metastasis/Origin 335 7.4
Diagnosis/Tumor Type 310 6.8
Staging/TNM 230 5.1
Sarcoma/Mesenchymal 214 4.7
Neuroendocrine 136 3.0
Margin/Resection 58 1.3
Second Opinion/Review 53 1.2
IHC/Biomarkers 28 0.6

Distribution of Consultation Categories Using the 13-Category Classification System

Distribution of Consultation Categories Using the 13-Category Classification System
NoteInterpretation

This chart shows what types of questions pathologists most commonly ask each other about. The most frequent categories reveal the diagnostic areas where peer consultation is most needed. A single consultation can be tagged with multiple categories, so the total exceeds the number of consultations.

11.3.1 Multi-Tag Analysis

Each consultation may match multiple categories. The tag count distribution and co-occurrence patterns reveal the complexity of consultation questions.

NoteInterpretation

This heatmap shows which consultation topics tend to appear together in the same case. Darker cells indicate category pairs that frequently co-occur. For example, a high co-occurrence between “Diagnosis/Tumor Type” and “IHC/Biomarkers” would mean these questions are often asked together, reflecting the integrated nature of diagnostic workup in those cases.

11.3.2 Category x Responder Heatmap

Distribution of consultation categories across the most active responders.

11.3.3 Category Turnaround Time

Turnaround time distribution by consultation category reveals differences in diagnostic complexity.

Turnaround Time Distribution by Consultation Category, Ordered by Median TAT (Capped at 72 Hours)
NoteInterpretation

This boxplot compares how quickly different types of consultations are answered. Categories at the top take longer on average (higher median turnaround time), which may reflect greater diagnostic complexity. The box shows the middle 50% of response times, and dots beyond the whiskers are unusually fast or slow outliers. Categories with wider boxes have more variable response times.

11.4 Organ System Analysis

Distribution of cases by Organ system (derived from ‘Paylasilan Vakalar’).

Top 20 Organs by Unique Case Count
Organ n
Kolon 197
Mide 152
Mide Biopsi 119
Kolon Biopsi 112
Tiroid 106
Meme 87
Akciger 75
Deri 68
Uterus 62
Lenf Nodu 50
Yumusak Doku Lipom Haric 49
Bobrek 40
Kemik Kuretaj 39
Karaciger Tru-Cut 37
Endometrium 36
Larinks 36
Hazir Parafin Blok 31
Kemik Iligi 26
Vulva,Labia Biopsi 26
Burun 24

11.5 Integrated Analysis

11.5.1 Organ vs. Subspecialty

Which organs are consulted by which subspecialties?